vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-10 17:14:36 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	.github/workflows/build.yml
#	.github/workflows/close-issue.yml
#	.github/workflows/server.yml
#	AUTHORS
#	CMakeLists.txt
#	Makefile
#	README.md
#	cmake/llama.pc.in
#	common/CMakeLists.txt
#	docs/build.md
#	examples/batched.swift/Sources/main.swift
#	examples/llama.swiftui/llama.cpp.swift/LibLlama.swift
#	examples/llava/CMakeLists.txt
#	examples/llava/clip.h
#	examples/run/run.cpp
#	examples/server/README.md
#	ggml/CMakeLists.txt
#	ggml/src/ggml-cuda/CMakeLists.txt
#	ggml/src/ggml-hip/CMakeLists.txt
#	ggml/src/ggml-musa/CMakeLists.txt
#	scripts/sync-ggml.last
#	tests/CMakeLists.txt
#	tests/test-backend-ops.cpp
#	tests/test-chat-template.cpp
#	tests/test-grammar-integration.cpp
#	tests/test-json-schema-to-grammar.cpp
This commit is contained in:
Concedo 2025-02-07 00:52:31 +08:00
commit db6db9dff9
142 changed files with 74261 additions and 10014 deletions
Download patch file Download diff file Expand all files Collapse all files

17
common/arg.cpp
View file

@ -1466,15 +1466,28 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
{"--list-devices"},
"print list of available devices and exit",
[](common_params &) {
printf("Available devices:\n");
std::vector<ggml_backend_dev_t> rpc_devices;
std::vector<ggml_backend_dev_t> all_devices;
for (size_t i = 0; i < ggml_backend_dev_count(); ++i) {
auto * dev = ggml_backend_dev_get(i);
if (ggml_backend_dev_type(dev) == GGML_BACKEND_DEVICE_TYPE_GPU) {
ggml_backend_reg_t reg = ggml_backend_dev_backend_reg(dev);
if (ggml_backend_reg_name(reg) == std::string("RPC")) {
rpc_devices.push_back(dev);
} else {
all_devices.push_back(dev);
}
}
}
// insert RPC devices in front
all_devices.insert(all_devices.begin(), rpc_devices.begin(), rpc_devices.end());
printf("Available devices:\n");
for (size_t i = 0; i < all_devices.size(); ++i) {
auto * dev = all_devices[i];
size_t free, total;
ggml_backend_dev_memory(dev, &free, &total);
printf(" %s: %s (%zu MiB, %zu MiB free)\n", ggml_backend_dev_name(dev), ggml_backend_dev_description(dev), total / 1024 / 1024, free / 1024 / 1024);
}
}
exit(0);
}
));

209
common/chat-template.hpp
View file

@ -33,6 +33,29 @@ struct chat_template_caps {
bool requires_typed_content = false;
};
struct chat_template_inputs {
nlohmann::ordered_json messages;
nlohmann::ordered_json tools;
bool add_generation_prompt = true;
nlohmann::ordered_json extra_context;
std::chrono::system_clock::time_point now = std::chrono::system_clock::now();
};
struct chat_template_options {
bool apply_polyfills = true;
bool use_bos_token = true;
bool use_eos_token = true;
bool define_strftime_now = true;
bool polyfill_tools = true;
bool polyfill_tool_call_examples = true;
bool polyfill_tool_calls = true;
bool polyfill_tool_responses = true;
bool polyfill_system_role = true;
bool polyfill_object_arguments = true;
bool polyfill_typed_content = true;
};
class chat_template {
private:
@ -41,6 +64,7 @@ class chat_template {
std::string bos_token_;
std::string eos_token_;
std::shared_ptr<minja::TemplateNode> template_root_;
std::string tool_call_example_;
std::string try_raw_render(
const nlohmann::ordered_json & messages,
@ -49,7 +73,18 @@ class chat_template {
const nlohmann::ordered_json & extra_context = nlohmann::ordered_json()) const
{
try {
auto prompt = apply(messages, tools, add_generation_prompt, extra_context, /* adjust_inputs= */ false);
chat_template_inputs inputs;
inputs.messages = messages;
inputs.tools = tools;
inputs.add_generation_prompt = add_generation_prompt;
inputs.extra_context = extra_context;
// Use fixed date for tests
inputs.now = std::chrono::system_clock::from_time_t(0);
chat_template_options opts;
opts.apply_polyfills = false;
auto prompt = apply(inputs, opts);
// fprintf(stderr, "try_raw_render: %s\n", prompt.c_str());
return prompt;
} catch (const std::exception & e) {
@ -176,6 +211,58 @@ class chat_template {
caps_.supports_tool_responses = contains(out, "Some response!");
caps_.supports_tool_call_id = contains(out, "call_911_");
}
try {
if (!caps_.supports_tools) {
const json user_msg {
{"role", "user"},
{"content", "Hey"},
};
const json args {
{"arg1", "some_value"},
};
const json tool_call_msg {
{"role", "assistant"},
{"content", nullptr},
{"tool_calls", json::array({
{
// TODO: detect if requires numerical id or fixed length == 6 like Nemo
{"id", "call_1___"},
{"type", "function"},
{"function", {
{"name", "tool_name"},
{"arguments", (caps_.requires_object_arguments ? args : json(minja::Value(args).dump(-1, /* to_json= */ true)))},
}},
},
})},
};
std::string prefix, full;
{
chat_template_inputs inputs;
inputs.messages = json::array({user_msg});
inputs.add_generation_prompt = true;
prefix = apply(inputs);
}
{
chat_template_inputs inputs;
inputs.messages = json::array({user_msg, tool_call_msg});
inputs.add_generation_prompt = false;
full = apply(inputs);
}
if (full.find(prefix) != 0) {
if (prefix.rfind(eos_token_) == prefix.size() - eos_token_.size()) {
prefix = prefix.substr(0, prefix.size() - eos_token_.size());
}
}
if (full.find(prefix) != 0) {
fprintf(stderr, "Failed to infer a tool call example (possible template bug)\n");
}
tool_call_example_ = full.substr(prefix.size());
}
} catch (const std::exception & e) {
fprintf(stderr, "Failed to generate tool call example: %s\n", e.what());
}
}
const std::string & source() const { return source_; }
@ -183,28 +270,72 @@ class chat_template {
const std::string & eos_token() const { return eos_token_; }
const chat_template_caps & original_caps() const { return caps_; }
// Deprecated, please use the form with chat_template_inputs and chat_template_options
std::string apply(
const nlohmann::ordered_json & messages,
const nlohmann::ordered_json & tools,
bool add_generation_prompt,
const nlohmann::ordered_json & extra_context = nlohmann::ordered_json(),
bool adjust_inputs = true) const
bool apply_polyfills = true)
{
fprintf(stderr, "[%s] Deprecated!\n", __func__);
chat_template_inputs inputs;
inputs.messages = messages;
inputs.tools = tools;
inputs.add_generation_prompt = add_generation_prompt;
inputs.extra_context = extra_context;
inputs.now = std::chrono::system_clock::now();
chat_template_options opts;
opts.apply_polyfills = apply_polyfills;
return apply(inputs, opts);
}
std::string apply(
const chat_template_inputs & inputs,
const chat_template_options & opts = chat_template_options()) const
{
json actual_messages;
auto needs_adjustments = adjust_inputs && (false
|| !caps_.supports_system_role
|| !caps_.supports_tools
|| !caps_.supports_tool_responses
|| !caps_.supports_tool_calls
|| caps_.requires_object_arguments
|| caps_.requires_typed_content
auto has_tools = inputs.tools.is_array() && !inputs.tools.empty();
auto has_tool_calls = false;
auto has_tool_responses = false;
auto has_string_content = false;
for (const auto & message : inputs.messages) {
if (message.contains("tool_calls") && !message["tool_calls"].is_null()) {
has_tool_calls = true;
}
if (message.contains("role") && message["role"] == "tool") {
has_tool_responses = true;
}
if (message.contains("content") && message["content"].is_string()) {
has_string_content = true;
}
}
auto polyfill_system_role = opts.polyfill_system_role && !caps_.supports_system_role;
auto polyfill_tools = opts.polyfill_tools && has_tools && !caps_.supports_tools;
auto polyfill_tool_call_example = polyfill_tools && opts.polyfill_tool_call_examples;
auto polyfill_tool_calls = opts.polyfill_tool_calls && has_tool_calls && !caps_.supports_tool_calls;
auto polyfill_tool_responses = opts.polyfill_tool_responses && has_tool_responses && !caps_.supports_tool_responses;
auto polyfill_object_arguments = opts.polyfill_object_arguments && has_tool_calls && caps_.requires_object_arguments;
auto polyfill_typed_content = opts.polyfill_typed_content && has_string_content && caps_.requires_typed_content;
auto needs_polyfills = opts.apply_polyfills && (false
|| polyfill_system_role
|| polyfill_tools
|| polyfill_tool_calls
|| polyfill_tool_responses
|| polyfill_object_arguments
|| polyfill_typed_content
);
if (needs_adjustments) {
if (needs_polyfills) {
actual_messages = json::array();
auto add_message = [&](const json & msg) {
if (caps_.requires_typed_content && msg.contains("content") && !msg.at("content").is_null() && msg.at("content").is_string()) {
if (polyfill_typed_content && msg.contains("content") && !msg.at("content").is_null() && msg.at("content").is_string()) {
actual_messages.push_back({
{"role", msg.at("role")},
{"content", {{
@ -227,9 +358,17 @@ class chat_template {
pending_system.clear();
}
};
auto needs_tools_in_system = !tools.is_null() && tools.size() > 0 && !caps_.supports_tools;
for (const auto & message_ : needs_tools_in_system ? add_system(messages, "Available tools: " + tools.dump(2)) : messages) {
json adjusted_messages;
if (polyfill_tools) {
adjusted_messages = add_system(inputs.messages,
"You can call any of the following tools to satisfy the user's requests: " + minja::Value(inputs.tools).dump(2, /* to_json= */ true) +
(!polyfill_tool_call_example || tool_call_example_.empty() ? "" : "\n\nExample tool call syntax:\n\n" + tool_call_example_));
} else {
adjusted_messages = inputs.messages;
}
for (const auto & message_ : adjusted_messages) {
auto message = message_;
if (!message.contains("role") || !message.contains("content")) {
throw std::runtime_error("message must have 'role' and 'content' fields: " + message.dump());
@ -237,7 +376,7 @@ class chat_template {
std::string role = message.at("role");
if (message.contains("tool_calls")) {
if (caps_.requires_object_arguments || !caps_.supports_tool_calls) {
if (polyfill_object_arguments || polyfill_tool_calls) {
for (auto & tool_call : message.at("tool_calls")) {
if (tool_call["type"] == "function") {
auto & function = tool_call.at("function");
@ -252,7 +391,7 @@ class chat_template {
}
}
}
if (!caps_.supports_tool_calls) {
if (polyfill_tool_calls) {
auto content = message.at("content");
auto tool_calls = json::array();
for (const auto & tool_call : message.at("tool_calls")) {
@ -279,7 +418,7 @@ class chat_template {
message.erase("tool_calls");
}
}
if (!caps_.supports_tool_responses && role == "tool") {
if (polyfill_tool_responses && role == "tool") {
message["role"] = "user";
auto obj = json {
{"tool_response", {
@ -296,7 +435,7 @@ class chat_template {
message.erase("name");
}
if (!message["content"].is_null() && !caps_.supports_system_role) {
if (!message["content"].is_null() && polyfill_system_role) {
std::string content = message.at("content");
if (role == "system") {
if (!pending_system.empty()) pending_system += "\n";
@ -315,28 +454,36 @@ class chat_template {
}
add_message(message);
}
if (!caps_.supports_system_role) {
flush_sys();
}
} else {
actual_messages = messages;
actual_messages = inputs.messages;
}
auto context = minja::Context::make(json({
{"messages", actual_messages},
{"add_generation_prompt", add_generation_prompt},
{"bos_token", bos_token_},
{"eos_token", eos_token_},
{"add_generation_prompt", inputs.add_generation_prompt},
}));
context->set("bos_token", opts.use_bos_token ? bos_token_ : "");
context->set("eos_token", opts.use_eos_token ? eos_token_ : "");
if (opts.define_strftime_now) {
auto now = inputs.now;
context->set("strftime_now", Value::callable([now](const std::shared_ptr<minja::Context> &, minja::ArgumentsValue & args) {
args.expectArgs("strftime_now", {1, 1}, {0, 0});
auto format = args.args[0].get<std::string>();
if (!tools.is_null()) {
auto tools_val = minja::Value(tools);
context->set("tools", tools_val);
auto time = std::chrono::system_clock::to_time_t(now);
auto local_time = *std::localtime(&time);
std::ostringstream ss;
ss << std::put_time(&local_time, format.c_str());
return ss.str();
}));
}
if (!extra_context.is_null()) {
for (auto & kv : extra_context.items()) {
minja::Value val(kv.value());
context->set(kv.key(), val);
if (!inputs.tools.is_null()) {
context->set("tools", minja::Value(inputs.tools));
}
if (!inputs.extra_context.is_null()) {
for (auto & kv : inputs.extra_context.items()) {
context->set(kv.key(), minja::Value(kv.value()));
}
}
@ -353,7 +500,7 @@ class chat_template {
std::string existing_system = messages_with_system.at(0).at("content");
messages_with_system[0] = json {
{"role", "system"},
{"content", existing_system + "\n" + system_prompt},
{"content", existing_system + "\n\n" + system_prompt},
};
} else {
messages_with_system.insert(messages_with_system.begin(), json {

129
common/chat.cpp
View file

@ -16,6 +16,7 @@ std::string common_chat_format_name(common_chat_format format) {
case COMMON_CHAT_FORMAT_FUNCTIONARY_V3_2: return "Functionary v3.2";
case COMMON_CHAT_FORMAT_FUNCTIONARY_V3_1_LLAMA_3_1: return "Functionary v3.1 Llama 3.1";
case COMMON_CHAT_FORMAT_HERMES_2_PRO: return "Hermes 2 Pro";
case COMMON_CHAT_FORMAT_COMMAND_R7B: return "Command R7B";
default:
throw std::runtime_error("Unknown chat format");
}
@ -162,6 +163,28 @@ static void foreach_function(const json & tools, const std::function<void(const
}
}
static std::string apply(
const common_chat_template & tmpl,
const nlohmann::ordered_json & messages,
const nlohmann::ordered_json & tools,
bool add_generation_prompt,
const nlohmann::ordered_json & extra_context = nlohmann::ordered_json())
{
minja::chat_template_inputs tmpl_inputs;
tmpl_inputs.messages = messages;
tmpl_inputs.tools = tools;
tmpl_inputs.add_generation_prompt = add_generation_prompt;
tmpl_inputs.extra_context = extra_context;
// TODO: add flag to control date/time, if only for testing purposes.
// tmpl_inputs.now = std::chrono::system_clock::now();
minja::chat_template_options tmpl_opts;
tmpl_opts.use_bos_token = false;
tmpl_opts.use_eos_token = false;
return tmpl.apply(tmpl_inputs, tmpl_opts);
}
static common_chat_params common_chat_params_init_generic(const common_chat_template & tmpl, const struct common_chat_inputs & inputs) {
common_chat_params data;
@ -243,7 +266,7 @@ static common_chat_params common_chat_params_init_generic(const common_chat_temp
inputs.messages,
"Respond in JSON format, either with `tool_call` (a request to call tools) or with `response` reply to the user's request");
data.prompt = tmpl.apply(tweaked_messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, tweaked_messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_GENERIC;
return data;
}
@ -309,7 +332,7 @@ static common_chat_params common_chat_params_init_mistral_nemo(const common_chat
builder.add_rule("root", "\"[TOOL_CALLS]\" " + builder.add_schema("tool_calls", schema));
}, grammar_options);
data.grammar_triggers.push_back({"[TOOL_CALLS]", /* .at_start = */ true});
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_MISTRAL_NEMO;
return data;
}
@ -317,6 +340,79 @@ static common_chat_msg common_chat_parse_mistral_nemo(const std::string & input)
return parse_prefixed_json_tool_call_array(input, "[TOOL_CALLS]");
}
static common_chat_params common_chat_params_init_command_r7b(const common_chat_template & tmpl, const struct common_chat_inputs & inputs) {
common_chat_params data;
data.grammar_lazy = inputs.tool_choice != "required";
data.grammar = build_grammar([&](const common_grammar_builder & builder) {
auto schemas = json::array();
foreach_function(inputs.tools, [&](const json & tool) {
const auto & function = tool["function"];
schemas.push_back({
{"type", "object"},
{"properties", {
{"tool_call_id", {
{"type", "string"},
// Command-R's template expects an integer string.
{"pattern", "^[0-9]{1,10}$"},
}},
{"tool_name", {
{"type", "string"},
{"const", function["name"]},
}},
{"parameters", function["parameters"]},
}},
{"required", json::array({"tool_call_id", "tool_name", "parameters"})},
});
});
auto schema = json {
{"type", "array"},
{"items", schemas.size() == 1 ? schemas[0] : json {{"anyOf", schemas}}},
{"minItems", 1},
};
if (!inputs.parallel_tool_calls) {
schema["maxItems"] = 1;
}
builder.add_rule("root", "\"<|START_ACTION|>\" " + builder.add_schema("tool_calls", schema) + " \"<|END_ACTION|>\"");
}, grammar_options);
data.grammar_triggers.push_back({"<|START_ACTION|>", /* .at_start = */ false});
data.preserved_tokens = {
"<|START_RESPONSE|>",
"<|END_RESPONSE|>",
"<|START_THINKING|>",
"<|END_THINKING|>",
"<|END_ACTION|>",
};
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_COMMAND_R7B;
return data;
}
static common_chat_msg common_chat_parse_command_r7b(const std::string & input) {
static std::regex response_regex("<\\|START_RESPONSE\\|>([\\s\\S\\n\\r]*?)<\\|END_RESPONSE\\|>");
static std::regex thought_action_regex("<\\|START_THINKING\\|>([\\s\\S\\n\\r]*?)<\\|END_THINKING\\|><\\|START_ACTION\\|>([\\s\\S\\n\\r]*?)<\\|END_ACTION\\|>");
std::smatch match;
common_chat_msg result;
result.role = "assistant";
if (std::regex_match(input, match, response_regex)) {
result.content = match[1].str();
} else if (std::regex_match(input, match, thought_action_regex)) {
result.tool_plan = match[1].str();
auto actions_str = match[2].str();
auto actions = json::parse(actions_str);
for (const auto & action : actions) {
result.tool_calls.push_back({
/* .name = */ action["tool_name"],
/* .arguments = */ action["parameters"].dump(),
/* .id = */ action["tool_call_id"],
});
}
} else {
LOG_ERR("Failed to parse command_r output");
result.content = input;
}
return result;
}
static void expect_tool_parameters(const std::string & name, const json & parameters, const std::vector<std::string> & expected_properties) {
if (!parameters.is_object() || !parameters.contains("type") || parameters["type"] != "object" || !parameters.contains("properties") || !parameters.contains("required")) {
throw std::runtime_error("Parameters of tool " + name + " must be an object w/ required properties");
@ -403,7 +499,7 @@ static common_chat_params common_chat_params_init_llama_3_1_tool_calls(const com
builder.add_rule("root", string_join(tool_rules, " | "));
}, grammar_options);
data.additional_stops.push_back("<|eom_id|>");
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt, {
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt, {
{"tools_in_user_message", false},
{"builtin_tools", builtin_tools.empty() ? json() : builtin_tools},
});
@ -462,9 +558,14 @@ static common_chat_params common_chat_params_init_deepseek_r1(const common_chat_
"\"<｜tool▁call▁begin｜>function<｜tool▁sep｜>" + name + "\\n```json\\n\" " + args_rule + " \"```<｜tool▁call▁end｜>\""));
});
data.grammar_triggers.push_back({"<｜tool▁calls▁begin｜>", /* .at_start = */ false});
data.preserved_tokens = {
"<｜tool▁sep｜>",
"<｜tool▁call▁end｜>",
};
builder.add_rule("root", "\"<｜tool▁calls▁begin｜>\" (" + string_join(tool_rules, " | ") + ")" + (inputs.parallel_tool_calls ? "*" : "") + " space");
}, grammar_options);
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
auto prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = prompt;
data.format = COMMON_CHAT_FORMAT_DEEPSEEK_R1;
return data;
}
@ -478,10 +579,10 @@ static common_chat_msg common_chat_parse_deepseek_r1(const std::string & input)
static common_chat_params common_chat_params_init_firefunction_v2(const common_chat_template & tmpl, const struct common_chat_inputs & inputs) {
fprintf(stderr, "%s\n", __func__);
common_chat_params data;
data.prompt = tmpl.apply(inputs.messages, /* tools= */ nullptr, inputs.add_generation_prompt, {
data.prompt = apply(tmpl, inputs.messages, /* tools= */ nullptr, inputs.add_generation_prompt, {
{"datetime", "Jan 29 2025 13:00:00 GMT"},
{"functions", json(inputs.tools.empty() ? "" : inputs.tools.dump(2))},
}, /* adjust_inputs= */ false);
});
if (!inputs.tools.is_null() && !inputs.tools.empty()) {
data.grammar_lazy = inputs.tool_choice != "required";
data.grammar = build_grammar([&](const common_grammar_builder & builder) {
@ -525,7 +626,7 @@ static common_chat_params common_chat_params_init_functionary_v3_2(const common_
// >>>all\nlet's call functions>>>fn1\n{"arg1": 1...}\n>>>fn2\n{"arg1": 1...}...
// Using ">>>f1\n", ">>>f2\n"... as trigger words for the grammar
common_chat_params data;
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_FUNCTIONARY_V3_2;
if (!inputs.tools.is_null() && !inputs.tools.empty()) {
data.grammar_lazy = inputs.tool_choice != "required";
@ -652,7 +753,7 @@ static common_chat_params common_chat_params_init_functionary_v3_1_llama_3_1(con
data.grammar_triggers.push_back({"<function=", /* .at_start = */ false});
}, grammar_options);
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
// TODO: if (has_raw_python)
data.format = COMMON_CHAT_FORMAT_FUNCTIONARY_V3_1_LLAMA_3_1;
return data;
@ -704,11 +805,10 @@ static common_chat_params common_chat_params_init_hermes_2_pro(const common_chat
auto tool_call = "\"<tool_call>\" space " + builder.add_rule("tool_call", string_join(tool_rules, " | ")) + " \"</tool_call>\" space";
builder.add_rule("root", inputs.parallel_tool_calls ? "(" + tool_call + ")+" : tool_call);
data.grammar_triggers.push_back({"<tool_call>", /* .at_start = */ false});
// Not really a trigger but need to print this special token to get a successful parse.
data.grammar_triggers.push_back({"</tool_call>", /* .at_start = */ false});
data.preserved_tokens = { "</tool_call>" };
}, grammar_options);
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_HERMES_2_PRO;
return data;
}
@ -769,7 +869,7 @@ static common_chat_msg common_chat_parse_hermes_2_pro(const std::string & input)
static common_chat_params common_chat_params_init_without_tools(const common_chat_template & tmpl, const struct common_chat_inputs & inputs) {
common_chat_params data;
data.prompt = tmpl.apply(inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.prompt = apply(tmpl, inputs.messages, inputs.tools.empty() ? json() : inputs.tools, inputs.add_generation_prompt);
data.format = COMMON_CHAT_FORMAT_CONTENT_ONLY;
data.grammar_lazy = false;
if (!inputs.json_schema.is_null()) {
@ -822,6 +922,9 @@ common_chat_params common_chat_params_init(const common_chat_template & tmpl, co
if (src.find("[TOOL_CALLS]") != std::string::npos) {
return common_chat_params_init_mistral_nemo(tmpl, inputs);
}
if (src.find("<|END_THINKING|><|START_ACTION|>") != std::string::npos) {
return common_chat_params_init_command_r7b(tmpl, inputs);
}
return common_chat_params_init_generic(tmpl, inputs);
}
@ -855,6 +958,8 @@ common_chat_msg common_chat_parse(const std::string & input, common_chat_format
return common_chat_parse_hermes_2_pro(input);
case COMMON_CHAT_FORMAT_FIREFUNCTION_V2:
return common_chat_parse_firefunction_v2(input);
case COMMON_CHAT_FORMAT_COMMAND_R7B:
return common_chat_parse_command_r7b(input);
default:
throw std::runtime_error("Unsupported format: " + common_chat_format_name(format));
}

2
common/chat.hpp
View file

@ -32,6 +32,7 @@ enum common_chat_format {
COMMON_CHAT_FORMAT_FUNCTIONARY_V3_2,
COMMON_CHAT_FORMAT_FUNCTIONARY_V3_1_LLAMA_3_1,
COMMON_CHAT_FORMAT_HERMES_2_PRO,
COMMON_CHAT_FORMAT_COMMAND_R7B,
COMMON_CHAT_FORMAT_COUNT, // Not a format, just the # formats
};
@ -42,6 +43,7 @@ struct common_chat_params {
std::string grammar;
bool grammar_lazy = false;
std::vector<common_grammar_trigger> grammar_triggers;
std::vector<std::string> preserved_tokens;
std::vector<std::string> additional_stops;
};

37
common/common.cpp
View file

@ -1871,11 +1871,19 @@ std::string common_chat_format_example(const common_chat_template & tmpl, bool u
return common_chat_apply_template(tmpl, msgs, true, use_jinja);
}
#define CHATML_TEMPLATE_SRC \
"{%- for message in messages -%}\n" \
" {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>\n' -}}\n" \
"{%- endfor -%}\n" \
"{%- if add_generation_prompt -%}\n" \
" {{- '<|im_start|>assistant\n' -}}\n" \
"{%- endif -%}"
common_chat_templates common_chat_templates_from_model(const struct llama_model * model, const std::string & chat_template_override)
{
auto vocab = llama_model_get_vocab(model);
std::string default_template_src = chat_template_override;
std::string template_tool_use_src = chat_template_override;
std::string default_template_src;
std::string template_tool_use_src;
bool has_explicit_template = !chat_template_override.empty();
if (chat_template_override.empty()) {
auto str = llama_model_chat_template(model, /* name */ nullptr);
@ -1888,21 +1896,17 @@ common_chat_templates common_chat_templates_from_model(const struct llama_model
template_tool_use_src = str;
has_explicit_template = true;
}
} else {
default_template_src = chat_template_override;
}
if (default_template_src.empty() || default_template_src == "chatml") {
if (!template_tool_use_src.empty()) {
default_template_src = template_tool_use_src;
} else {
default_template_src = R"(
{%- for message in messages -%}
{{- "<|im_start|>" + message.role + "\n" + message.content + "<|im_end|>\n" -}}
{%- endfor -%}
{%- if add_generation_prompt -%}
{{- "<|im_start|>assistant\n" -}}
{%- endif -%}
)";
default_template_src = CHATML_TEMPLATE_SRC;
}
}
auto vocab = llama_model_get_vocab(model);
const auto get_token = [&](llama_token token, const char * name, const char * jinja_variable_name) {
if (token == LLAMA_TOKEN_NULL) {
if (default_template_src.find(jinja_variable_name) != std::string::npos
@ -1916,13 +1920,22 @@ common_chat_templates common_chat_templates_from_model(const struct llama_model
};
auto token_bos = get_token(llama_vocab_bos(vocab), "BOS", "bos_token");
auto token_eos = get_token(llama_vocab_eos(vocab), "EOS", "eos_token");
try {
return {
has_explicit_template,
std::make_unique<minja::chat_template>(default_template_src, token_bos, token_eos),
template_tool_use_src.empty()
? nullptr
: std::make_unique<minja::chat_template>(template_tool_use_src, token_bos, token_eos)
: std::make_unique<minja::chat_template>(template_tool_use_src, token_bos, token_eos),
};
} catch (const std::exception & e) {
LOG_ERR("%s: failed to parse chat template: %s\n", __func__, e.what());
return {
has_explicit_template,
std::make_unique<minja::chat_template>(CHATML_TEMPLATE_SRC, token_bos, token_eos),
nullptr,
};
}
}
//

3
common/common.h
View file

@ -4,6 +4,7 @@
#include "llama-cpp.h"
#include <set>
#include <string>
#include <vector>
#include <sstream>
@ -159,6 +160,7 @@ struct common_params_sampling {
bool grammar_lazy = false;
std::vector<common_grammar_trigger> grammar_trigger_words; // optional trigger words to trigger lazy grammar
std::vector<llama_token> grammar_trigger_tokens; // optional trigger tokens to trigger lazy grammar and print trigger special tokens.
std::set<llama_token> preserved_tokens;
std::vector<llama_logit_bias> logit_bias; // logit biases to apply
@ -617,6 +619,7 @@ struct common_chat_msg {
std::string role;
std::string content;
std::vector<common_tool_call> tool_calls;
std::string tool_plan = "";
};
// Check if the template supplied via "--chat-template" is supported or not. Returns true if it's valid

9
common/json-schema-to-grammar.cpp
View file

@ -991,7 +991,14 @@ public:
}
};
std::string json_schema_to_grammar(const json & schema) {
std::string json_schema_to_grammar(const json & schema, bool force_gbnf) {
#ifdef LLAMA_USE_LLGUIDANCE
if (!force_gbnf) {
return "%llguidance {}\nstart: %json " + schema.dump();
}
#else
(void)force_gbnf;
#endif // LLAMA_USE_LLGUIDANCE
return build_grammar([&](const common_grammar_builder & callbacks) {
auto copy = schema;
callbacks.resolve_refs(copy);

3
common/json-schema-to-grammar.h
View file

@ -5,7 +5,8 @@
#define JSON_ASSERT GGML_ASSERT
#include "json.hpp"
std::string json_schema_to_grammar(const nlohmann::ordered_json & schema);
std::string json_schema_to_grammar(const nlohmann::ordered_json & schema,
bool force_gbnf = false);
struct common_grammar_builder {
std::function<std::string(const std::string &, const std::string &)> add_rule;

270
common/llguidance.cpp Normal file
View file

@ -0,0 +1,270 @@
#include "sampling.h"
#include "log.h"
#ifdef LLAMA_USE_LLGUIDANCE
# include "llguidance.h"
# include <cmath>
struct llama_sampler_llg {
const llama_vocab * vocab;
std::string grammar_kind;
std::string grammar_data;
LlgTokenizer * tokenizer;
LlgConstraint * grammar;
LlgMaskResult llg_res;
bool has_llg_res;
};
static LlgConstraint * llama_sampler_llg_new(LlgTokenizer * tokenizer, const char * grammar_kind,
const char * grammar_data) {
LlgConstraintInit cinit;
llg_constraint_init_set_defaults(&cinit, tokenizer);
const char * log_level = getenv("LLGUIDANCE_LOG_LEVEL");
if (log_level && *log_level) {
cinit.log_stderr_level = atoi(log_level);
}
auto c = llg_new_constraint_any(&cinit, grammar_kind, grammar_data);
if (llg_get_error(c)) {
LOG_ERR("llg error: %s\n", llg_get_error(c));
llg_free_constraint(c);
return nullptr;
}
return c;
}
static const char * llama_sampler_llg_name(const llama_sampler * /*smpl*/) {
return "llguidance";
}
static void llama_sampler_llg_accept_impl(llama_sampler * smpl, llama_token token) {
auto * ctx = (llama_sampler_llg *) smpl->ctx;
if (ctx->grammar) {
LlgCommitResult res;
llg_commit_token(ctx->grammar, token, &res);
ctx->has_llg_res = false;
}
}
static void llama_sampler_llg_apply(llama_sampler * smpl, llama_token_data_array * cur_p) {
auto * ctx = (llama_sampler_llg *) smpl->ctx;
if (ctx->grammar) {
if (!ctx->has_llg_res) {
if (llg_compute_mask(ctx->grammar, &ctx->llg_res) == 0) {
ctx->has_llg_res = true;
} else {
LOG_ERR("llg error: %s\n", llg_get_error(ctx->grammar));
llg_free_constraint(ctx->grammar);
ctx->grammar = nullptr;
}
}
if (ctx->has_llg_res) {
if (ctx->llg_res.is_stop) {
for (size_t i = 0; i < cur_p->size; ++i) {
if (!llama_vocab_is_eog(ctx->vocab, cur_p->data[i].id)) {
cur_p->data[i].logit = -INFINITY;
}
}
} else {
const uint32_t * mask = ctx->llg_res.sample_mask;
for (size_t i = 0; i < cur_p->size; ++i) {
auto token = cur_p->data[i].id;
if ((mask[token / 32] & (1 << (token % 32))) == 0) {
cur_p->data[i].logit = -INFINITY;
}
}
}
}
}
}
static void llama_sampler_llg_reset(llama_sampler * smpl) {
auto * ctx = (llama_sampler_llg *) smpl->ctx;
if (!ctx->grammar) {
return;
}
auto * grammar_new = llama_sampler_llg_new(ctx->tokenizer, ctx->grammar_kind.c_str(), ctx->grammar_data.c_str());
llg_free_constraint(ctx->grammar);
ctx->grammar = grammar_new;
ctx->has_llg_res = false;
}
static llama_sampler * llama_sampler_llg_clone(const llama_sampler * smpl) {
const auto * ctx = (const llama_sampler_llg *) smpl->ctx;
auto * result = llama_sampler_init_llg(ctx->vocab, nullptr, nullptr);
// copy the state
{
auto * result_ctx = (llama_sampler_llg *) result->ctx;
if (ctx->grammar) {
result_ctx->grammar_kind = ctx->grammar_kind;
result_ctx->grammar_data = ctx->grammar_data;
result_ctx->grammar = llg_clone_constraint(ctx->grammar);
result_ctx->tokenizer = llg_clone_tokenizer(ctx->tokenizer);
}
}
return result;
}
static void llama_sampler_llg_free(llama_sampler * smpl) {
const auto * ctx = (llama_sampler_llg *) smpl->ctx;
if (ctx->grammar) {
llg_free_constraint(ctx->grammar);
llg_free_tokenizer(ctx->tokenizer);
}
delete ctx;
}
static llama_sampler_i llama_sampler_llg_i = {
/* .name = */ llama_sampler_llg_name,
/* .accept = */ llama_sampler_llg_accept_impl,
/* .apply = */ llama_sampler_llg_apply,
/* .reset = */ llama_sampler_llg_reset,
/* .clone = */ llama_sampler_llg_clone,
/* .free = */ llama_sampler_llg_free,
};
static size_t llama_sampler_llg_tokenize_fn(const void * user_data, const uint8_t * bytes, size_t bytes_len,
uint32_t * output_tokens, size_t output_tokens_len) {
const llama_vocab * vocab = (const llama_vocab *) user_data;
int r = 0;
try {
r = llama_tokenize(vocab, (const char *) bytes, bytes_len, (int32_t *) output_tokens, output_tokens_len, false,
true);
} catch (const std::exception & e) {
GGML_ABORT("llama_tokenize failed: %s\n", e.what());
}
if (r < 0) {
return -r;
}
return r;
}
static LlgTokenizer * llama_sampler_llg_new_tokenizer(const llama_vocab * vocab) {
// TODO store the tokenizer in the vocab somehow
static const llama_vocab * vocab_cache;
static LlgTokenizer * tokenizer_cache;
if (vocab_cache == vocab) {
return llg_clone_tokenizer(tokenizer_cache);
}
auto tok_eos = llama_vocab_eot(vocab);
if (tok_eos == LLAMA_TOKEN_NULL) {
tok_eos = llama_vocab_eos(vocab);
}
size_t vocab_size = llama_vocab_n_tokens(vocab);
auto token_lens = new uint32_t[vocab_size];
// we typically have ~7 bytes per token; let's go on the safe side here
auto token_bytes_size = vocab_size * 16 + 1024 * 1024;
auto token_bytes = new uint8_t[token_bytes_size];
size_t offset = 0;
for (size_t i = 0; i < vocab_size; i++) {
size_t max_token = 1024;
if (token_bytes_size - offset < max_token) {
GGML_ABORT("token_bytes buffer too small\n");
}
llama_token token = i;
auto dp = (char *) token_bytes + offset;
auto size = llama_detokenize(vocab, &token, 1, dp, max_token, false, false);
if (size < 0) {
GGML_ABORT("llama_detokenize failed\n");
}
if (size == 0) {
size = llama_detokenize(vocab, &token, 1, dp + 1, max_token - 1, false, true);
if (size < 0) {
GGML_ABORT("llama_detokenize failed\n");
}
if (size != 0) {
*dp = '\xff'; // special token prefix marker
size += 1;
}
}
token_lens[i] = size;
offset += size;
}
LlgTokenizerInit tinit = {
/* .vocab_size = */ (uint32_t) vocab_size,
/* .tok_eos = */ (uint32_t) tok_eos,
/* .token_lens = */ token_lens,
/* .token_bytes = */ token_bytes,
/* .tokenizer_json = */ nullptr,
/* .tokenize_assumes_string = */ true,
/* .tokenize_fn = */ llama_sampler_llg_tokenize_fn,
/* .use_approximate_greedy_tokenize_fn = */ false,
/* .tokenize_user_data = */ vocab,
};
char error_buffer[1024];
LlgTokenizer * tokenizer = llg_new_tokenizer(&tinit, error_buffer, sizeof(error_buffer));
delete[] token_bytes;
delete[] token_lens;
if (tokenizer == nullptr) {
LOG_ERR("llg tokenizer error: %s\n", error_buffer);
return tokenizer;
}
if (tokenizer_cache) {
llg_free_tokenizer(tokenizer_cache);
}
vocab_cache = vocab;
tokenizer_cache = tokenizer;
return llg_clone_tokenizer(tokenizer_cache);
}
llama_sampler * llama_sampler_init_llg(const llama_vocab * vocab, const char * grammar_kind,
const char * grammar_data) {
auto * ctx = new llama_sampler_llg;
if (grammar_kind != nullptr && grammar_kind[0] != '\0') {
auto tokenizer = llama_sampler_llg_new_tokenizer(vocab);
*ctx = {
/* .vocab = */ vocab,
/* .grammar_kind = */ grammar_kind,
/* .grammar_data = */ grammar_data,
/* .tokenizer = */ tokenizer,
/* .grammar = */ llama_sampler_llg_new(tokenizer, grammar_kind, grammar_data),
/* .llg_res = */ {},
/* .has_llg_res = */ false,
};
} else {
*ctx = {
/* .vocab = */ vocab,
/* .grammar_kind = */ {},
/* .grammar_data = */ {},
/* .tokenizer = */ nullptr,
/* .grammar = */ nullptr,
/* .llg_res = */ {},
/* .has_llg_res = */ false,
};
}
return new llama_sampler{
/* .iface = */ &llama_sampler_llg_i,
/* .ctx = */ ctx,
};
}
#else
llama_sampler * llama_sampler_init_llg(const llama_vocab *, const char *, const char *) {
LOG_WRN("llguidance (cmake -DLLAMA_LLGUIDANCE=ON) is not enabled");
return nullptr;
}
#endif // LLAMA_USE_LLGUIDANCE

10
common/log.cpp
View file

@ -14,16 +14,6 @@ void common_log_set_verbosity_thold(int verbosity) {
common_log_verbosity_thold = verbosity;
}
#define LOG_COL_DEFAULT "\033[0m"
#define LOG_COL_BOLD "\033[1m"
#define LOG_COL_RED "\033[31m"
#define LOG_COL_GREEN "\033[32m"
#define LOG_COL_YELLOW "\033[33m"
#define LOG_COL_BLUE "\033[34m"
#define LOG_COL_MAGENTA "\033[35m"
#define LOG_COL_CYAN "\033[36m"
#define LOG_COL_WHITE "\033[37m"
static int64_t t_us() {
return std::chrono::duration_cast<std::chrono::microseconds>(std::chrono::system_clock::now().time_since_epoch()).count();
}

10
common/log.h
View file

@ -2,6 +2,16 @@
#include "ggml.h" // for ggml_log_level
#define LOG_COL_DEFAULT "\033[0m"
#define LOG_COL_BOLD "\033[1m"
#define LOG_COL_RED "\033[31m"
#define LOG_COL_GREEN "\033[32m"
#define LOG_COL_YELLOW "\033[33m"
#define LOG_COL_BLUE "\033[34m"
#define LOG_COL_MAGENTA "\033[35m"
#define LOG_COL_CYAN "\033[36m"
#define LOG_COL_WHITE "\033[37m"
#ifndef __GNUC__
# define LOG_ATTRIBUTE_FORMAT(...)
#elif defined(__MINGW32__)

55
common/minja.hpp
View file

@ -693,7 +693,7 @@ enum SpaceHandling { Keep, Strip, StripSpaces, StripNewline };
class TemplateToken {
public:
enum class Type { Text, Expression, If, Else, Elif, EndIf, For, EndFor, Generation, EndGeneration, Set, EndSet, Comment, Macro, EndMacro, Filter, EndFilter };
enum class Type { Text, Expression, If, Else, Elif, EndIf, For, EndFor, Generation, EndGeneration, Set, EndSet, Comment, Macro, EndMacro, Filter, EndFilter, Break, Continue };
static std::string typeToString(Type t) {
switch (t) {
@ -714,6 +714,8 @@ public:
case Type::EndFilter: return "endfilter";
case Type::Generation: return "generation";
case Type::EndGeneration: return "endgeneration";
case Type::Break: return "break";
case Type::Continue: return "continue";
}
return "Unknown";
}
@ -815,6 +817,22 @@ struct CommentTemplateToken : public TemplateToken {
CommentTemplateToken(const Location & location, SpaceHandling pre, SpaceHandling post, const std::string& t) : TemplateToken(Type::Comment, location, pre, post), text(t) {}
};
enum class LoopControlType { Break, Continue };
class LoopControlException : public std::runtime_error {
public:
LoopControlType control_type;
LoopControlException(const std::string & message, LoopControlType control_type) : std::runtime_error(message), control_type(control_type) {}
LoopControlException(LoopControlType control_type)
: std::runtime_error((control_type == LoopControlType::Continue ? "continue" : "break") + std::string(" outside of a loop")),
control_type(control_type) {}
};
struct LoopControlTemplateToken : public TemplateToken {
LoopControlType control_type;
LoopControlTemplateToken(const Location & location, SpaceHandling pre, SpaceHandling post, LoopControlType control_type) : TemplateToken(Type::Break, location, pre, post), control_type(control_type) {}
};
class TemplateNode {
Location location_;
protected:
@ -825,6 +843,12 @@ public:
void render(std::ostringstream & out, const std::shared_ptr<Context> & context) const {
try {
do_render(out, context);
} catch (const LoopControlException & e) {
// TODO: make stack creation lazy. Only needed if it was thrown outside of a loop.
std::ostringstream err;
err << e.what();
if (location_.source) err << error_location_suffix(*location_.source, location_.pos);
throw LoopControlException(err.str(), e.control_type);
} catch (const std::exception & e) {
std::ostringstream err;
err << e.what();
@ -897,6 +921,15 @@ public:
}
};
class LoopControlNode : public TemplateNode {
LoopControlType control_type_;
public:
LoopControlNode(const Location & location, LoopControlType control_type) : TemplateNode(location), control_type_(control_type) {}
void do_render(std::ostringstream &, const std::shared_ptr<Context> &) const override {
throw LoopControlException(control_type_);
}
};
class ForNode : public TemplateNode {
std::vector<std::string> var_names;
std::shared_ptr<Expression> iterable;
@ -961,7 +994,12 @@ public:
loop.set("last", i == (n - 1));
loop.set("previtem", i > 0 ? filtered_items.at(i - 1) : Value());
loop.set("nextitem", i < n - 1 ? filtered_items.at(i + 1) : Value());
try {
body->render(out, loop_context);
} catch (const LoopControlException & e) {
if (e.control_type == LoopControlType::Break) break;
if (e.control_type == LoopControlType::Continue) continue;
}
}
}
};
@ -2156,10 +2194,10 @@ private:
}
TemplateTokenVector tokenize() {
static std::regex comment_tok(R"(\{#([-~]?)(.*?)([-~]?)#\})");
static std::regex comment_tok(R"(\{#([-~]?)([\s\S\r\n]*?)([-~]?)#\})");
static std::regex expr_open_regex(R"(\{\{([-~])?)");
static std::regex block_open_regex(R"(^\{%([-~])?[\s\n\r]*)");
static std::regex block_keyword_tok(R"((if|else|elif|endif|for|endfor|generation|endgeneration|set|endset|block|endblock|macro|endmacro|filter|endfilter)\b)");
static std::regex block_keyword_tok(R"((if|else|elif|endif|for|endfor|generation|endgeneration|set|endset|block|endblock|macro|endmacro|filter|endfilter|break|continue)\b)");
static std::regex non_text_open_regex(R"(\{\{|\{%|\{#)");
static std::regex expr_close_regex(R"([\s\n\r]*([-~])?\}\})");
static std::regex block_close_regex(R"([\s\n\r]*([-~])?%\})");
@ -2291,6 +2329,9 @@ private:
} else if (keyword == "endfilter") {
auto post_space = parseBlockClose();
tokens.push_back(std::make_unique<EndFilterTemplateToken>(location, pre_space, post_space));
} else if (keyword == "break" || keyword == "continue") {
auto post_space = parseBlockClose();
tokens.push_back(std::make_unique<LoopControlTemplateToken>(location, pre_space, post_space, keyword == "break" ? LoopControlType::Break : LoopControlType::Continue));
} else {
throw std::runtime_error("Unexpected block: " + keyword);
}
@ -2414,6 +2455,8 @@ private:
children.emplace_back(std::make_shared<FilterNode>(token->location, std::move(filter_token->filter), std::move(body)));
} else if (dynamic_cast<CommentTemplateToken*>(token.get())) {
// Ignore comments
} else if (auto ctrl_token = dynamic_cast<LoopControlTemplateToken*>(token.get())) {
children.emplace_back(std::make_shared<LoopControlNode>(token->location, ctrl_token->control_type));
} else if (dynamic_cast<EndForTemplateToken*>(token.get())
|| dynamic_cast<EndSetTemplateToken*>(token.get())
|| dynamic_cast<EndMacroTemplateToken*>(token.get())
@ -2572,6 +2615,7 @@ inline std::shared_ptr<Context> Context::builtins() {
}));
globals.set("join", simple_function("join", { "items", "d" }, [](const std::shared_ptr<Context> &, Value & args) {
auto do_join = [](Value & items, const std::string & sep) {
if (!items.is_array()) throw std::runtime_error("object is not iterable: " + items.dump());
std::ostringstream oss;
auto first = true;
for (size_t i = 0, n = items.size(); i < n; ++i) {
@ -2652,6 +2696,10 @@ inline std::shared_ptr<Context> Context::builtins() {
return Value::callable([=](const std::shared_ptr<Context> & context, ArgumentsValue & args) {
args.expectArgs(is_select ? "select" : "reject", {2, (std::numeric_limits<size_t>::max)()}, {0, 0});
auto & items = args.args[0];
if (items.is_null())
return Value::array();
if (!items.is_array()) throw std::runtime_error("object is not iterable: " + items.dump());
auto filter_fn = context->get(args.args[1]);
if (filter_fn.is_null()) throw std::runtime_error("Undefined filter: " + args.args[1].dump());
@ -2729,6 +2777,7 @@ inline std::shared_ptr<Context> Context::builtins() {
auto & items = args.args[0];
if (items.is_null())
return Value::array();
if (!items.is_array()) throw std::runtime_error("object is not iterable: " + items.dump());
auto attr_name = args.args[1].get<std::string>();
bool has_test = false;

20
common/sampling.cpp
View file

@ -156,13 +156,25 @@ struct common_sampler * common_sampler_init(const struct llama_model * model, co
for (const auto & str : params.grammar_trigger_words) {
trigger_words.push_back(str.word.c_str());
}
auto * result = new common_sampler {
/* .params = */ params,
/* .grmr = */ params.grammar_lazy
struct llama_sampler * grmr;
if (params.grammar.compare(0, 11, "%llguidance") == 0) {
#ifdef LLAMA_USE_LLGUIDANCE
grmr = llama_sampler_init_llg(vocab, "lark", params.grammar.c_str());
#else
GGML_ABORT("llguidance (cmake -DLLAMA_LLGUIDANCE=ON) is not enabled");
#endif // LLAMA_USE_LLGUIDANCE
} else {
grmr = params.grammar_lazy
? llama_sampler_init_grammar_lazy(vocab, params.grammar.c_str(), "root",
trigger_words.data(), trigger_words.size(),
params.grammar_trigger_tokens.data(), params.grammar_trigger_tokens.size())
: llama_sampler_init_grammar(vocab, params.grammar.c_str(), "root"),
: llama_sampler_init_grammar(vocab, params.grammar.c_str(), "root");
}
auto * result = new common_sampler {
/* .params = */ params,
/* .grmr = */ grmr,
/* .chain = */ llama_sampler_chain_init(lparams),
/* .prev = */ ring_buffer<llama_token>(std::max(32, params.n_prev)),
/* .cur = */ {},

3
common/sampling.h
View file

@ -102,3 +102,6 @@ std::string common_sampler_type_to_str(enum common_sampler_type cnstr);
std::vector<enum common_sampler_type> common_sampler_types_from_names(const std::vector<std::string> & names, bool allow_alt_names);
std::vector<enum common_sampler_type> common_sampler_types_from_chars(const std::string & chars);
llama_sampler * llama_sampler_init_llg(const llama_vocab * vocab,
const char * grammar_kind, const char * grammar_data);

58
convert_hf_to_gguf.py
View file

@ -648,7 +648,7 @@ class Model:
if chkhsh == "7967bfa498ade6b757b064f31e964dddbb80f8f9a4d68d4ba7998fcf281c531a":
# ref: https://huggingface.co/jinaai/jina-embeddings-v2-base-code
res = "jina-v2-code"
if chkhsh == "b6e8e1518dc4305be2fe39c313ed643381c4da5db34a98f6a04c093f8afbe99b":
if chkhsh == "b6e8e1518dc4305be2fe39c313ed643381c4da5db34a98f6a04c093f8afbe99b" or chkhsh == "81d72c7348a9f0ebe86f23298d37debe0a5e71149e29bd283904c02262b27516":
# ref: https://huggingface.co/THUDM/glm-4-9b-chat
res = "chatglm-bpe"
if chkhsh == "7fc505bd3104ca1083b150b17d088b59534ede9bde81f0dd2090967d7fe52cee":
@ -4513,7 +4513,7 @@ class JaisModel(Model):
self.gguf_writer.add_max_alibi_bias(self.max_alibi_bias)
@Model.register("ChatGLMModel", "ChatGLMForConditionalGeneration")
@Model.register("GlmForCausalLM", "ChatGLMModel", "ChatGLMForConditionalGeneration")
class ChatGLMModel(Model):
model_arch = gguf.MODEL_ARCH.CHATGLM
@ -4619,47 +4619,15 @@ class ChatGLMModel(Model):
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(dir_model, trust_remote_code=True)
vocab_size = hparams["padded_vocab_size"]
vocab_size = hparams.get("padded_vocab_size",hparams["vocab_size"])
assert max(tokenizer.get_vocab().values()) < vocab_size
tokpre = self.get_vocab_base_pre(tokenizer)
merges = []
vocab = {}
mergeable_ranks = tokenizer.mergeable_ranks
for token, rank in mergeable_ranks.items():
vocab[ChatGLMModel.token_bytes_to_string(token)] = rank
if len(token) == 1:
continue
merged = ChatGLMModel.bpe(mergeable_ranks, token, max_rank=rank)
assert len(merged) >= 2 and len(merged) <= 7
merges.append(' '.join(map(ChatGLMModel.token_bytes_to_string, merged)))
# for this kind of tokenizer, added_vocab is not a subset of vocab, so they need to be combined
added_vocab = tokenizer.get_added_vocab()
reverse_vocab = {id_ : encoded_tok for encoded_tok, id_ in {**vocab, **added_vocab}.items()}
for i in range(vocab_size):
if i not in reverse_vocab:
tokens.append(f"[PAD{i}]")
toktypes.append(gguf.TokenType.UNUSED)
elif reverse_vocab[i] in added_vocab:
tokens.append(reverse_vocab[i])
if tokenizer.added_tokens_decoder[i].special:
toktypes.append(gguf.TokenType.CONTROL)
else:
toktypes.append(gguf.TokenType.USER_DEFINED)
else:
tokens.append(reverse_vocab[i])
toktypes.append(gguf.TokenType.NORMAL)
tokens, toktypes, tokpre = self.get_vocab_base()
self.gguf_writer.add_tokenizer_model("gpt2")
self.gguf_writer.add_tokenizer_pre(tokpre)
self.gguf_writer.add_token_list(tokens)
self.gguf_writer.add_token_types(toktypes)
special_vocab = gguf.SpecialVocab(dir_model, load_merges=False)
special_vocab.merges = merges
special_vocab = gguf.SpecialVocab(self.dir_model, load_merges=True)
# only add special tokens when they were not already loaded from config.json
special_vocab._set_special_token("eos", tokenizer.get_added_vocab()["<|endoftext|>"])
special_vocab._set_special_token("eot", tokenizer.get_added_vocab()["<|user|>"])
@ -4670,16 +4638,20 @@ class ChatGLMModel(Model):
def set_gguf_parameters(self):
n_embed = self.hparams.get("hidden_size", self.hparams.get("n_embed"))
n_head = self.hparams.get("n_head", self.hparams.get("num_attention_heads"))
n_head_kv = self.hparams.get("multi_query_group_num", n_head)
n_head_kv = self.hparams.get("multi_query_group_num", self.hparams.get("num_key_value_heads", n_head))
self.gguf_writer.add_context_length(self.hparams.get("seq_length", n_embed))
self.gguf_writer.add_embedding_length(n_embed)
self.gguf_writer.add_feed_forward_length(self.hparams.get("ffn_hidden_size", 4 * n_embed))
self.gguf_writer.add_block_count(self.hparams["num_layers"])
self.gguf_writer.add_feed_forward_length(self.hparams.get("ffn_hidden_size", self.hparams.get("intermediate_size", 4 * n_embed)))
self.gguf_writer.add_block_count(self.hparams.get("num_layers", self.hparams["num_hidden_layers"]))
self.gguf_writer.add_head_count(n_head)
self.gguf_writer.add_head_count_kv(n_head_kv)
self.gguf_writer.add_layer_norm_rms_eps(self.hparams["layernorm_epsilon"])
self.gguf_writer.add_layer_norm_rms_eps(self.hparams.get("layernorm_epsilon",1e-5))
self.gguf_writer.add_file_type(self.ftype)
self.gguf_writer.add_rope_dimension_count(64)
if "attention_dim" in self.hparams:
rope_dim = self.hparams["attention_dim"]
else:
rope_dim = self.hparams["hidden_size"] // self.hparams["num_attention_heads"]
self.gguf_writer.add_rope_dimension_count(int(rope_dim * self.hparams.get("partial_rotary_factor", 0.5)))
self.gguf_writer.add_add_bos_token(False)
rope_freq = 10000
if "rope_ratio" in self.hparams:
@ -4689,7 +4661,7 @@ class ChatGLMModel(Model):
def modify_tensors(self, data_torch: Tensor, name: str, bid: int | None) -> Iterable[tuple[str, Tensor]]:
del bid # unused
if name.endswith(".rotary_pos_emb.inv_freq"):
if name.endswith(".rotary_pos_emb.inv_freq") or name.startswith("model.vision."):
return []
name = name.removeprefix("transformer.")

317
docs/cuda-fedora.md
View file

@ -1,317 +0,0 @@
# Setting Up CUDA on Fedora
In this guide we setup [Nvidia CUDA](https://docs.nvidia.com/cuda/) in a toolbox container. This guide is applicable for:
- [Fedora Workstation](https://fedoraproject.org/workstation/)
- [Atomic Desktops for Fedora](https://fedoraproject.org/atomic-desktops/)
- [Fedora Spins](https://fedoraproject.org/spins)
- [Other Distributions](https://containertoolbx.org/distros/), including `Red Hat Enterprise Linux >= 8.`, `Arch Linux`, and `Ubuntu`.
## Table of Contents
- [Prerequisites](#prerequisites)
- [Monitoring NVIDIA CUDA Repositories](#monitoring-nvidia-cuda-repositories)
- [Using the Fedora 39 CUDA Repository](#using-the-fedora-39-cuda-repository)
- [Creating a Fedora Toolbox Environment](#creating-a-fedora-toolbox-environment)
- [Installing Essential Development Tools](#installing-essential-development-tools)
- [Adding the CUDA Repository](#adding-the-cuda-repository)
- [Installing `nvidia-driver-libs`](#installing-nvidia-driver-libs)
- [Manually Resolving Package Conflicts](#manually-resolving-package-conflicts)
- [Finalizing the Installation of `nvidia-driver-libs`](#finalizing-the-installation-of-nvidia-driver-libs)
- [Installing the CUDA Meta-Package](#installing-the-cuda-meta-package)
- [Configuring the Environment](#configuring-the-environment)
- [Verifying the Installation](#verifying-the-installation)
- [Conclusion](#conclusion)
- [Troubleshooting](#troubleshooting)
- [Additional Notes](#additional-notes)
- [References](#references)
## Prerequisites
- **Toolbox Installed on the Host System** `Fedora Silverblue` and `Fedora Workstation` both have toolbox by default, other distributions may need to install the [toolbox package](https://containertoolbx.org/install/).
- **NVIDIA Drivers and Graphics Card installed on Host System (optional)** To run CUDA program, such as `llama.cpp`, the host should be setup to access your NVIDIA hardware. Fedora Hosts can use the [RPM Fusion Repository](https://rpmfusion.org/Howto/NVIDIA).
- **Internet connectivity** to download packages.
### Monitoring NVIDIA CUDA Repositories
Before proceeding, it is advisable to check if NVIDIA has updated their CUDA repositories for your Fedora version. NVIDIA's repositories can be found at:
- [Fedora 40 CUDA Repository](https://developer.download.nvidia.com/compute/cuda/repos/fedora40/x86_64/)
- [Fedora 41 CUDA Repository](https://developer.download.nvidia.com/compute/cuda/repos/fedora41/x86_64/)
As of the latest update, these repositories do not contain the `cuda` meta-package or are missing essential components.
### Using the Fedora 39 CUDA Repository
Since the newer repositories are incomplete, we'll use the Fedora 39 repository:
- [Fedora 39 CUDA Repository](https://developer.download.nvidia.com/compute/cuda/repos/fedora39/x86_64/)
**Note:** Fedora 39 is no longer maintained, so we recommend using a toolbox environment to prevent system conflicts.
## Creating a Fedora Toolbox Environment
This guide focuses on Fedora hosts, but with small adjustments, it can work for other hosts. Using a Fedora 39 toolbox allows us to install the necessary packages without affecting the host system.
**Note:** Toolbox is available for other systems, and even without Toolbox, it is possible to use Podman or Docker.
We do not recommend installing on the host system, as Fedora 39 is out-of-maintenance, and instead you should upgrade to a maintained version of Fedora for your host.
1. **Create a Fedora 39 Toolbox:**
```bash
toolbox create --image registry.fedoraproject.org/fedora-toolbox:39 --container fedora-toolbox-39-cuda
```
2. **Enter the Toolbox:**
```bash
toolbox enter --container fedora-toolbox-39-cuda
```
Inside the toolbox, you have root privileges and can install packages without affecting the host system.
## Installing Essential Development Tools
1. **Synchronize the DNF Package Manager:**
```bash
sudo dnf distro-sync
```
2. **Install the Default Text Editor (Optional):**
```bash
sudo dnf install vim-default-editor --allowerasing
```
The `--allowerasing` flag resolves any package conflicts.
3. **Install Development Tools and Libraries:**
```bash
sudo dnf install @c-development @development-tools cmake
```
This installs essential packages for compiling software, including `gcc`, `make`, and other development headers.
## Adding the CUDA Repository
Add the NVIDIA CUDA repository to your DNF configuration:
```bash
sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora39/x86_64/cuda-fedora39.repo
```
After adding the repository, synchronize the package manager again:
```bash
sudo dnf distro-sync
```
## Installing `nvidia-driver-libs`
Attempt to install `nvidia-driver-libs`:
```bash
sudo dnf install nvidia-driver-libs
```
**Explanation:**
- `nvidia-driver-libs` contains necessary NVIDIA driver libraries required by CUDA.
- This step might fail due to conflicts with existing NVIDIA drivers on the host system.
## Manually Resolving Package Conflicts
If the installation fails due to conflicts, we'll manually download and install the required packages, excluding conflicting files.
### 1. Download the `nvidia-driver-libs` RPM
```bash
sudo dnf download --arch x86_64 nvidia-driver-libs
```
You should see a file similar to:
```
nvidia-driver-libs-560.35.05-1.fc39.x86_64.rpm
```
### 2. Attempt to Install the RPM
```bash
sudo dnf install nvidia-driver-libs-560.35.05-1.fc39.x86_64.rpm
```
**Expected Error:**
Installation may fail with errors pointing to conflicts with `egl-gbm` and `egl-wayland`.
**Note: It is important to carefully read the error messages to identify the exact paths that need to be excluded.**
### 3. Download Dependencies
```bash
sudo dnf download --arch x86_64 egl-gbm egl-wayland
```
### 4. Install `egl-gbm` with Excluded Paths
Exclude conflicting files during installation:
```bash
sudo rpm --install --verbose --hash \
--excludepath=/usr/lib64/libnvidia-egl-gbm.so.1.1.2 \
--excludepath=/usr/share/egl/egl_external_platform.d/15_nvidia_gbm.json \
egl-gbm-1.1.2^20240919gitb24587d-3.fc39.x86_64.rpm
```
**Explanation:**
- The `--excludepath` option skips installing files that conflict with existing files.
- Adjust the paths based on the error messages you receive.
### 5. Install `egl-wayland` with Excluded Paths
```bash
sudo rpm --install --verbose --hash \
--excludepath=/usr/share/egl/egl_external_platform.d/10_nvidia_wayland.json \
egl-wayland-1.1.17^20241118giteeb29e1-5.fc39.x86_64.rpm
```
### 6. Install `nvidia-driver-libs` with Excluded Paths
```bash
sudo rpm --install --verbose --hash \
--excludepath=/usr/share/glvnd/egl_vendor.d/10_nvidia.json \
--excludepath=/usr/share/nvidia/nvoptix.bin \
nvidia-driver-libs-560.35.05-1.fc39.x86_64.rpm
```
**Note:**
- Replace the paths with the ones causing conflicts in your installation if they differ.
- The `--verbose` and `--hash` options provide detailed output during installation.
## Finalizing the Installation of `nvidia-driver-libs`
After manually installing the dependencies, run:
```bash
sudo dnf install nvidia-driver-libs
```
You should receive a message indicating the package is already installed:
```
Package nvidia-driver-libs-3:560.35.05-1.fc39.x86_64 is already installed.
Dependencies resolved.
Nothing to do.
Complete!
```
## Installing the CUDA Meta-Package
Now that the driver libraries are installed, proceed to install CUDA:
```bash
sudo dnf install cuda
```
This installs the CUDA toolkit and associated packages.
## Configuring the Environment
To use CUDA, add its binary directory to your system's `PATH`.
1. **Create a Profile Script:**
```bash
sudo sh -c 'echo "export PATH=\$PATH:/usr/local/cuda/bin" >> /etc/profile.d/cuda.sh'
```
**Explanation:**
- We add to `/etc/profile.d/` as the `/etc/` folder is unique to this particular container, and is not shared with other containers or the host system.
- The backslash `\` before `$PATH` ensures the variable is correctly written into the script.
2. **Make the Script Executable:**
```bash
sudo chmod +x /etc/profile.d/cuda.sh
```
3. **Source the Script to Update Your Environment:**
```bash
source /etc/profile.d/cuda.sh
```
**Note:** This command updates your current shell session with the new `PATH`. The `/etc/profile.d/cuda.sh` script ensures that the CUDA binaries are available in your `PATH` for all future sessions.
## Verifying the Installation
To confirm that CUDA is correctly installed and configured, check the version of the NVIDIA CUDA Compiler (`nvcc`):
```bash
nvcc --version
```
You should see output similar to:
```
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2024 NVIDIA Corporation
Built on Tue_Oct_29_23:50:19_PDT_2024
Cuda compilation tools, release 12.6, V12.6.85
Build cuda_12.6.r12.6/compiler.35059454_0
```
This output confirms that the CUDA compiler is accessible and indicates the installed version.
## Conclusion
You have successfully set up CUDA on Fedora within a toolbox environment using the Fedora 39 CUDA repository. By manually resolving package conflicts and configuring the environment, you can develop CUDA applications without affecting your host system.
## Troubleshooting
- **Installation Failures:**
- If you encounter errors during installation, carefully read the error messages. They often indicate conflicting files or missing dependencies.
- Use the `--excludepath` option with `rpm` to exclude conflicting files during manual installations.
- **Driver Conflicts:**
- Since the host system may already have NVIDIA drivers installed, conflicts can arise. Using the toolbox environment helps isolate these issues.
- **Environment Variables Not Set:**
- If `nvcc` is not found after installation, ensure that `/usr/local/cuda/bin` is in your `PATH`.
- Run `echo $PATH` to check if the path is included.
- Re-source the profile script or open a new terminal session.
## Additional Notes
- **Updating CUDA in the Future:**
- Keep an eye on the official NVIDIA repositories for updates to your Fedora version.
- When an updated repository becomes available, adjust your `dnf` configuration accordingly.
- **Building `llama.cpp`:**
- With CUDA installed, you can follow these [build instructions for `llama.cpp`](https://github.com/ggerganov/llama.cpp/blob/master/docs/build.md) to compile it with CUDA support.
- Ensure that any CUDA-specific build flags or paths are correctly set in your build configuration.
- **Using the Toolbox Environment:**
- The toolbox environment is isolated from your host system, which helps prevent conflicts.
- Remember that system files and configurations inside the toolbox are separate from the host. By default the home directory of the user is shared between the host and the toolbox.
---
**Disclaimer:** Manually installing and modifying system packages can lead to instability of the container. The above steps are provided as a guideline and may need adjustments based on your specific system configuration. Always back up important data before making significant system changes, especially as your home folder is writable and shared with he toolbox.
**Acknowledgments:** Special thanks to the Fedora community and NVIDIA documentation for providing resources that assisted in creating this guide.
## References
- [Fedora Toolbox Documentation](https://docs.fedoraproject.org/en-US/fedora-silverblue/toolbox/)
- [NVIDIA CUDA Installation Guide](https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html)
- [Podman Documentation](https://podman.io/get-started)
---

43
examples/llava/README-glmedge.md Normal file
View file

@ -0,0 +1,43 @@
# GLMV-EDGE
Currently this implementation supports [glm-edge-v-2b](https://huggingface.co/THUDM/glm-edge-v-2b) and [glm-edge-v-5b](https://huggingface.co/THUDM/glm-edge-v-5b).
## Usage
Build with cmake or run `make llama-llava-cli` to build it.
After building, run: `./llama-llava-cli` to see the usage. For example:
```sh
./llama-llava-cli -m model_path/ggml-model-f16.gguf --mmproj model_path/mmproj-model-f16.gguf --image img_path/image.jpg -p "<|system|>\n system prompt <image><|user|>\n prompt <|assistant|>\n"
```
**note**: A lower temperature like 0.1 is recommended for better quality. add `--temp 0.1` to the command to do so.
**note**: For GPU offloading ensure to use the `-ngl` flag just like usual
## GGUF conversion
1. Clone a GLMV-EDGE model ([2B](https://huggingface.co/THUDM/glm-edge-v-2b) or [5B](https://huggingface.co/THUDM/glm-edge-v-5b)). For example:
```sh
git clone https://huggingface.co/THUDM/glm-edge-v-5b or https://huggingface.co/THUDM/glm-edge-v-2b
```
2. Use `glmedge-surgery.py` to split the GLMV-EDGE model to LLM and multimodel projector constituents:
```sh
python ./examples/llava/glmedge-surgery.py -m ../model_path
```
4. Use `glmedge-convert-image-encoder-to-gguf.py` to convert the GLMV-EDGE image encoder to GGUF:
```sh
python ./examples/llava/glmedge-convert-image-encoder-to-gguf.py -m ../model_path --llava-projector ../model_path/glm.projector --output-dir ../model_path
```
5. Use `examples/convert_hf_to_gguf.py` to convert the LLM part of GLMV-EDGE to GGUF:
```sh
python convert_hf_to_gguf.py ../model_path
```
Now both the LLM part and the image encoder are in the `model_path` directory.

44
examples/llava/README-quantize.md Normal file
View file

@ -0,0 +1,44 @@
# Quantizing CLIP Visual Projector
This is the tool for quantizing the CLIP visual projector model. Quantization reduces the precision of the model's weights, which can significantly decrease the model size and improve inference speed, often with minimal impact on performance.
## Usage
To quantize a CLIP visual projector model, use the following command:
```sh
./bin/llama-llava-clip-quantize-cli /path/to/ggml-model-f32.gguf /path/to/ggml-model-quantized.gguf <type>
```
After the quantization, the visual projector can be used freely with the existing LLAVA cli (LLAVA, Qwen2VL, etc).
### Arguments
- `/path/to/ggml-model-f32.gguf`: The path to the input model file in FP32 or FP16 format.
- `/path/to/ggml-model-quantized.gguf`: The path where the quantized model will be saved.
- `<type>`: The quantization type to apply. This should be an integer corresponding to one of the quantization types defined in the `enum ggml_type`.
### Quantization Types
The following quantization types are supported, based on the `enum ggml_type` definition:
- `2` - `q4_0`: 4-bit quantization with a single scale value.
- `3` - `q4_1`: 4-bit quantization with a separate scale value for each block.
- `6` - `q5_0`: 5-bit quantization with a single scale value.
- `7` - `q5_1`: 5-bit quantization with a separate scale value for each block.
- `8` - `q8_0`: 8-bit quantization with a single scale value.
### Example
To quantize a model using the `q4_0` quantization type, you would run:
```sh
./bin/llama-llava-clip-quantize-cli /path/to/ggml-model-f32.gguf /path/to/ggml-model-quantized.gguf 2
```
This command will generate a quantized model at `/path/to/ggml-model-quantized.gguf` using the `q4_0` quantization method.
## Notes
- Quantization can lead to a loss in model accuracy, depending on the chosen quantization type. It is recommended to evaluate the quantized model's performance on your specific task to ensure it meets your requirements.
- The quantized model will typically be smaller in size and faster to run, making it more suitable for deployment in resource-constrained environments.

53
examples/llava/android/adb_run.sh
View file

@ -1,53 +0,0 @@
#!/bin/bash
model_dir="/Users/cxt/model/llm/mobileVLM/MobileVLM-1.7B_processed"
projector_name="mmproj-model-f16.gguf"
llama_name="ggml-model-q4_k.gguf"
img_dir="/Users/cxt/model/llm"
img_name="demo.jpg"
prompt="A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: <image>\nWho is the author of this book? \nAnswer the question using a single word or phrase. ASSISTANT:"
# img_name="cat.jpeg"
# prompt="A chat between a curious user and an artificial intelligence assistant. The assistant gives helpful, detailed, and polite answers to the user's questions. USER: <image>\nWhat is in the image? ASSISTANT:"
program_dir="build_64/bin"
binName="llama-llava-cli"
n_threads=4
deviceDir="/data/local/tmp"
saveDir="output"
if [ ! -d ${saveDir} ]; then
mkdir ${saveDir}
fi
function android_run() {
# # copy resource into device
# adb push ${model_dir}/${projector_name} ${deviceDir}/${projector_name}
# adb push ${model_dir}/${llama_name} ${deviceDir}/${llama_name}
adb push ${img_dir}/${img_name} ${deviceDir}/${img_name}
# copy program into device
adb push ${program_dir}/${binName} ${deviceDir}/${binName}
adb shell "chmod 0777 ${deviceDir}/${binName}"
# run
adb shell "echo cd ${deviceDir} ${deviceDir}/${binName} \
-m ${deviceDir}/${llama_name} \
--mmproj ${deviceDir}/${projector_name} \
-t ${n_threads} \
--image ${deviceDir}/${img_name} \
-p \"${prompt}\" \
> ${deviceDir}/${modelName}_${projector_name}_${n_threads}_${img_name}.txt"
adb shell "cd ${deviceDir}; pwd; ${deviceDir}/${binName} \
-m ${deviceDir}/${llama_name} \
--mmproj ${deviceDir}/${projector_name} \
-t ${n_threads} \
--image ${deviceDir}/${img_name} \
-p \"${prompt}\" \
>> ${deviceDir}/${modelName}_${projector_name}_${n_threads}_${img_name}.txt 2>&1"
adb pull ${deviceDir}/${modelName}_${projector_name}_${n_threads}_${img_name}.txt ${saveDir}
}
android_run
echo "android_run is Done!"

8
examples/llava/android/build_64.sh
View file

@ -1,8 +0,0 @@
#!/bin/bash
cmake ../../../../ \
-DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK/build/cmake/android.toolchain.cmake \
-DCMAKE_BUILD_TYPE=Release \
-DANDROID_ABI="arm64-v8a" \
-DANDROID_PLATFORM=android-23 $1
make -j4

59
examples/llava/clip-quantize-cli.cpp Normal file
View file

@ -0,0 +1,59 @@
#include "arg.h"
#include "base64.hpp"
#include "log.h"
#include "common.h"
#include "sampling.h"
#include "clip.h"
#include "llava.h"
#include "llama.h"
#include "ggml.h"
static void print_usage(int argc, char ** argv) {
(void) argc;
fprintf(stderr, "usage: %s /path/to/ggml-model-f32.gguf /path/to/ggml-model-quantized.gguf type\n", argv[0]);
fprintf(stderr, " type = 2 - q4_0\n");
fprintf(stderr, " type = 3 - q4_1\n");
fprintf(stderr, " type = 6 - q5_0\n");
fprintf(stderr, " type = 7 - q5_1\n");
fprintf(stderr, " type = 8 - q8_0\n");
}
int main(int argc, char ** argv) {
if (argc != 4) {
print_usage(argc, argv);
return 1;
}
const std::string fname_inp = argv[1];
const std::string fname_out = argv[2];
const int itype = atoi(argv[3]);
const int64_t t_main_start_us = ggml_time_us();
int64_t t_quantize_us = 0;
// load the model
{
const int64_t t_start_us = ggml_time_us();
if (!clip_model_quantize(fname_inp.c_str(), fname_out.c_str(), itype)) {
fprintf(stderr, "%s: failed to quantize model from '%s'\n", __func__, fname_inp.c_str());
return 1;
}
t_quantize_us = ggml_time_us() - t_start_us;
}
// report timing
{
const int64_t t_main_end_us = ggml_time_us();
printf("\n");
printf("%s: quantize time = %8.2f ms\n", __func__, t_quantize_us / 1000.0f);
printf("%s: total time = %8.2f ms\n", __func__, (t_main_end_us - t_main_start_us) / 1000.0f);
}
return 0;
}

118
examples/llava/clip.cpp
View file

@ -105,6 +105,7 @@ static std::string format(const char * fmt, ...) {
#define KEY_HAS_VIS_ENC "clip.has_vision_encoder"
#define KEY_HAS_LLAVA_PROJ "clip.has_llava_projector"
#define KEY_HAS_MINICPMV_PROJ "clip.has_minicpmv_projector"
#define KEY_HAS_GLM_PROJ "clip.has_glm_projector"
#define KEY_MINICPMV_VERSION "clip.minicpmv_version"
#define KEY_HAS_QWEN2VL_MERGER "clip.has_qwen2vl_merger"
#define KEY_USE_GELU "clip.use_gelu"
@ -163,6 +164,15 @@ static std::string format(const char * fmt, ...) {
#define TN_MINICPMV_ATTN "resampler.attn.%s.%s"
#define TN_MINICPMV_LN "resampler.ln_%s.%s"
#define TN_GLM_ADAPER_CONV "adapter.conv.%s"
#define TN_GLM_ADAPTER_LINEAR "adapter.linear.linear.%s"
#define TN_GLM_ADAPTER_NORM_1 "adapter.linear.norm1.%s"
#define TN_GLM_ADAPTER_D_H_2_4H "adapter.linear.dense_h_to_4h.%s"
#define TN_GLM_ADAPTER_GATE "adapter.linear.gate.%s"
#define TN_GLM_ADAPTER_D_4H_2_H "adapter.linear.dense_4h_to_h.%s"
#define TN_GLM_BOI_W "adapter.boi"
#define TN_GLM_EOI_W "adapter.eoi"
enum projector_type {
PROJECTOR_TYPE_MLP,
@ -170,6 +180,7 @@ enum projector_type {
PROJECTOR_TYPE_LDP,
PROJECTOR_TYPE_LDPV2,
PROJECTOR_TYPE_RESAMPLER,
PROJECTOR_TYPE_GLM_EDGE,
PROJECTOR_TYPE_MERGER,
PROJECTOR_TYPE_UNKNOWN,
};
@ -179,6 +190,7 @@ static std::map<projector_type, std::string> PROJECTOR_TYPE_NAMES = {
{ PROJECTOR_TYPE_LDP, "ldp" },
{ PROJECTOR_TYPE_LDPV2, "ldpv2"},
{ PROJECTOR_TYPE_RESAMPLER, "resampler"},
{ PROJECTOR_TYPE_GLM_EDGE, "adapter"},
{ PROJECTOR_TYPE_MERGER, "qwen2vl_merger"},
};
@ -503,6 +515,12 @@ struct clip_vision_model {
struct ggml_tensor * mm_4_w = NULL;
struct ggml_tensor * mm_4_b = NULL;
//GLMV-Edge projection
struct ggml_tensor * mm_model_adapter_conv_w;
struct ggml_tensor * mm_model_adapter_conv_b;
struct ggml_tensor * boi_w;
struct ggml_tensor * eoi_w;
// MobileVLM projection
struct ggml_tensor * mm_model_mlp_1_w;
struct ggml_tensor * mm_model_mlp_1_b;
@ -563,6 +581,7 @@ struct clip_ctx {
bool has_vision_encoder = false;
bool has_llava_projector = false;
bool has_minicpmv_projector = false;
bool has_glm_projector = false;
bool has_qwen2vl_merger = false;
int minicpmv_version = 2;
@ -641,7 +660,7 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32
const int batch_size = imgs->size;
if (ctx->has_llava_projector || ctx->has_minicpmv_projector) {
if (ctx->has_llava_projector || ctx->has_minicpmv_projector || ctx->has_glm_projector) {
GGML_ASSERT(batch_size == 1);
}
@ -737,8 +756,7 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32
}
// loop over layers
if (ctx->has_minicpmv_projector || ctx->has_qwen2vl_merger) {
// TODO: figure out why we doing thing in this way ???
if (ctx->has_minicpmv_projector || ctx->has_glm_projector || ctx->has_qwen2vl_merger) {
n_layer += 1;
}
for (int il = 0; il < n_layer - 1; il++) {
@ -1098,7 +1116,33 @@ static ggml_cgraph * clip_image_build_graph(clip_ctx * ctx, const clip_image_f32
GGML_ASSERT(false);
}
}
else if (ctx->proj_type == PROJECTOR_TYPE_MERGER) {
// glm projector
else if (ctx->has_glm_projector) {
if (ctx->proj_type == PROJECTOR_TYPE_GLM_EDGE) {
size_t gridsz = (size_t)sqrt(embeddings->ne[1]);
embeddings = ggml_cont(ctx0, ggml_permute(ctx0,embeddings,1,0,2,3));
embeddings = ggml_reshape_3d(ctx0, embeddings, gridsz, gridsz, embeddings->ne[1]);
embeddings = ggml_conv_2d(ctx0, model.mm_model_adapter_conv_w, embeddings, 2, 2, 0, 0, 1, 1);
embeddings = ggml_reshape_3d(ctx0, embeddings,embeddings->ne[0]*embeddings->ne[1] , embeddings->ne[2], batch_size);
embeddings = ggml_cont(ctx0, ggml_permute(ctx0,embeddings, 1, 0, 2, 3));
embeddings = ggml_add(ctx0, embeddings, model.mm_model_adapter_conv_b);
//GLU
{
embeddings = ggml_mul_mat(ctx0, model.mm_model_mlp_0_w, embeddings);
embeddings = ggml_norm(ctx0, embeddings, eps);
embeddings = ggml_add(ctx0, ggml_mul(ctx0, embeddings, model.mm_model_ln_q_w), model.mm_model_ln_q_b);
embeddings = ggml_gelu_inplace(ctx0, embeddings);
struct ggml_tensor * x = embeddings;
embeddings = ggml_mul_mat(ctx0, model.mm_model_mlp_2_w, embeddings);
x = ggml_mul_mat(ctx0, model.mm_model_mlp_1_w,x);
embeddings = ggml_silu_inplace(ctx0, embeddings);
embeddings = ggml_mul(ctx0, embeddings,x);
embeddings = ggml_mul_mat(ctx0, model.mm_model_mlp_3_w, embeddings);
}
} else {
GGML_ABORT("fatel error");
}
} else if (ctx->proj_type == PROJECTOR_TYPE_MERGER) {
embeddings = ggml_reshape_3d(ctx0, embeddings, hidden_size * 4, num_positions / 4, batch_size);
embeddings = ggml_mul_mat(ctx0, model.mm_0_w, embeddings);
@ -1296,6 +1340,11 @@ if(enable_gpu_clip)
new_clip->minicpmv_version = gguf_get_val_i32(ctx, idx);
}
idx = gguf_find_key(ctx, KEY_HAS_GLM_PROJ);
if (idx != -1) {
new_clip->has_glm_projector = gguf_get_val_bool(ctx, idx);
}
idx = gguf_find_key(ctx, KEY_HAS_QWEN2VL_MERGER);
if (idx != -1) {
new_clip->has_qwen2vl_merger = gguf_get_val_bool(ctx, idx);
@ -1320,6 +1369,7 @@ if(enable_gpu_clip)
LOG_INF("%s: vision_encoder: %d\n", __func__, new_clip->has_vision_encoder);
LOG_INF("%s: llava_projector: %d\n", __func__, new_clip->has_llava_projector);
LOG_INF("%s: minicpmv_projector: %d\n", __func__, new_clip->has_minicpmv_projector);
LOG_INF("%s: glm_projector: %d\n", __func__, new_clip->has_glm_projector);
LOG_INF("%s: model size: %.2f MB\n", __func__, model_size / 1024.0 / 1024.0);
LOG_INF("%s: metadata size: %.2f MB\n", __func__, ggml_get_mem_size(meta) / 1024.0 / 1024.0);
}
@ -1587,6 +1637,18 @@ if(enable_gpu_clip)
vision_model.mm_model_ln_post_w = get_tensor(new_clip->ctx_data, format(TN_MINICPMV_LN, "post", "weight"));
vision_model.mm_model_ln_post_b = get_tensor(new_clip->ctx_data, format(TN_MINICPMV_LN, "post", "bias"));
}
else if (new_clip->proj_type == PROJECTOR_TYPE_GLM_EDGE) {
vision_model.mm_model_adapter_conv_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPER_CONV, "weight"));
vision_model.mm_model_adapter_conv_b = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPER_CONV, "bias"));
vision_model.mm_model_mlp_0_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_LINEAR,"weight"));
vision_model.mm_model_ln_q_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_NORM_1,"weight"));
vision_model.mm_model_ln_q_b = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_NORM_1,"bias"));
vision_model.mm_model_mlp_1_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_D_H_2_4H,"weight"));
vision_model.mm_model_mlp_2_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_GATE,"weight"));
vision_model.mm_model_mlp_3_w = get_tensor(new_clip->ctx_data, format(TN_GLM_ADAPTER_D_4H_2_H,"weight"));
vision_model.boi_w = get_tensor(new_clip->ctx_data, TN_GLM_BOI_W);
vision_model.eoi_w = get_tensor(new_clip->ctx_data, TN_GLM_EOI_W);
}
else if (new_clip->proj_type == PROJECTOR_TYPE_MERGER) {
vision_model.mm_0_w = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 0, "weight"));
vision_model.mm_0_b = get_tensor(new_clip->ctx_data, format(TN_LLAVA_PROJ, 0, "bias"));
@ -2214,6 +2276,20 @@ bool clip_image_preprocess(struct clip_ctx * ctx, const clip_image_u8 * img, cli
return true;
}
if (ctx->has_glm_projector) {
res_imgs->size = 1;
res_imgs->data = new clip_image_f32[res_imgs->size];
clip_image_u8 resized_image;
int32_t sz=ctx->vision_model.hparams.image_size;
bicubic_resize(*img, resized_image,sz,sz);
clip_image_f32 * res = clip_image_f32_init();
//clip_image_save_to_bmp(resized_image, "resized.bmp");
normalize_image_u8_to_f32(&resized_image, res, ctx->image_mean, ctx->image_std);
res_imgs->data[0] = *res;
clip_image_f32_free(res);
return true;
}
bool pad_to_square = true;
if (!ctx->has_vision_encoder) {
LOG_ERR("This gguf file seems to have no vision encoder\n");
@ -2399,7 +2475,8 @@ void clip_free(clip_ctx * ctx) {
}
size_t clip_embd_nbytes(const struct clip_ctx * ctx) {
return clip_n_patches(ctx) * clip_n_mmproj_embd(ctx) * sizeof(float);
int extra_tokens = ctx->has_glm_projector ? 2 : 0;
return (clip_n_patches(ctx) + extra_tokens) * clip_n_mmproj_embd(ctx) * sizeof(float);
}
size_t clip_embd_nbytes_by_img(const struct clip_ctx * ctx, int img_h, int img_w) {
@ -2441,7 +2518,7 @@ int clip_n_patches_by_img(const struct clip_ctx * ctx, struct clip_image_f32 * i
int n_patches = (params.image_size / params.patch_size) * (params.image_size / params.patch_size);
if (ctx->proj_type == PROJECTOR_TYPE_LDP || ctx->proj_type == PROJECTOR_TYPE_LDPV2) {
if (ctx->proj_type == PROJECTOR_TYPE_LDP || ctx->proj_type == PROJECTOR_TYPE_LDPV2 || ctx->proj_type == PROJECTOR_TYPE_GLM_EDGE) {
n_patches /= 4;
} else if (ctx->proj_type == PROJECTOR_TYPE_RESAMPLER) {
if (ctx->minicpmv_version == 2) {
@ -2574,6 +2651,12 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
if (ctx->has_minicpmv_projector) {
GGML_ASSERT(batch_size == 1);
}
if (ctx->has_glm_projector) {
GGML_ASSERT(batch_size == 1);
ggml_tensor * boi = ctx->vision_model.boi_w;
ggml_backend_tensor_get(boi,vec,0,ggml_nbytes(boi));
vec = (float*)(vec+ggml_nelements(boi)); //offset for boi
}
// build the inference graph
ggml_cgraph * gf = clip_image_build_graph(ctx, imgs, ctx->load_image_size, true);
@ -2726,7 +2809,7 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
ggml_backend_tensor_set(positions, positions_data, 0, ggml_nbytes(positions));
free(positions_data);
{
if (!ctx->has_glm_projector) {
struct ggml_tensor * patches = ggml_graph_get_tensor(gf, "patches");
int* patches_data = (int*)malloc(ggml_nbytes(patches));
for (int i = 0; i < num_patches; i++) {
@ -2750,15 +2833,20 @@ bool clip_image_batch_encode(clip_ctx * ctx, const int n_threads, const clip_ima
// copy the embeddings to the location passed by the user
ggml_backend_tensor_get(embeddings, vec, 0, ggml_nbytes(embeddings));
if (ctx->has_glm_projector) {
//eoi
ggml_tensor * eoi = ctx->vision_model.eoi_w;
int offset = ggml_nelements(embeddings);
ggml_backend_tensor_get(eoi, vec+offset, 0, ggml_nbytes(eoi));
}
return true;
}
static bool avoid_problematic_indivisible = true;
bool clip_model_quantize(const char * fname_inp, const char * fname_out, const int itype) {
ggml_type type = GGML_TYPE_Q4_1;
assert(itype < GGML_TYPE_COUNT);
type = static_cast<ggml_type>(itype);
ggml_type type = static_cast<ggml_type>(itype);
auto * ctx_clip = clip_model_load(fname_inp, 2);
@ -2811,8 +2899,8 @@ bool clip_model_quantize(const char * fname_inp, const char * fname_out, const i
}
}
// quantize only 2D tensors
quantize &= (ggml_n_dims(cur) == 2);
// quantize only 2D tensors and bigger than block size
quantize &= (ggml_n_dims(cur) == 2) && cur->ne[0] > ggml_blck_size(type);
//kcpp fix: do not quantize certain tensors if they are indivisible!
if(avoid_problematic_indivisible)
@ -2921,6 +3009,9 @@ int clip_n_mmproj_embd(const struct clip_ctx * ctx) {
return 3584;
}
}
if (ctx->proj_type == PROJECTOR_TYPE_GLM_EDGE){
return ctx->vision_model.mm_model_mlp_3_w->ne[1];
}
if (ctx->proj_type == PROJECTOR_TYPE_MERGER) {
return ctx->vision_model.mm_1_b->ne[0];
}
@ -2936,6 +3027,9 @@ int clip_is_minicpmv(const struct clip_ctx * ctx) {
return 0;
}
bool clip_is_glm(const struct clip_ctx * ctx) {
return ctx->has_glm_projector;
}
bool clip_is_qwen2vl(const struct clip_ctx * ctx) {
return ctx->has_qwen2vl_merger;
}

3
examples/llava/clip.h
View file

@ -93,7 +93,10 @@ CLIP_API bool clip_is_qwen2vl(const struct clip_ctx * ctx);
CLIP_API bool clip_encode_float_image (struct clip_ctx * ctx, int n_threads, float * img, int h, int w, float * vec);
CLIP_API bool clip_is_glm(const struct clip_ctx * ctx);
CLIP_API void set_clip_uses_gpu(bool usegpu);
#ifdef __cplusplus
}
#endif

280
examples/llava/glmedge-convert-image-encoder-to-gguf.py Normal file
View file

@ -0,0 +1,280 @@
import argparse
import os
import json
import re
import torch
import numpy as np
from gguf import *
TEXT = "clip.text"
VISION = "clip.vision"
from transformers import SiglipVisionModel, SiglipVisionConfig
def k(raw_key: str, arch: str) -> str:
return raw_key.format(arch=arch)
def should_skip_tensor(name: str, has_text: bool, has_vision: bool, has_llava: bool) -> bool:
if name in (
"logit_scale",
"text_model.embeddings.position_ids",
"vision_model.embeddings.position_ids",
):
return True
if name in (
"vision_model.head.probe",
"vision_model.head.attention.in_proj_weight",
"vision_model.head.attention.in_proj_bias",
"vision_model.head.attention.out_proj.weight",
"vision_model.head.attention.out_proj.bias",
"vision_model.head.layernorm.weight",
"vision_model.head.layernorm.bias",
"vision_model.head.mlp.fc1.weight",
"vision_model.head.mlp.fc1.bias",
"vision_model.head.mlp.fc2.weight",
"vision_model.head.mlp.fc2.bias"
):
return True
if name.startswith("v") and not has_vision:
return True
if name.startswith("t") and not has_text:
return True
return False
def get_tensor_name(name: str) -> str:
if "projection" in name:
return name
if "mm_projector" in name:
name = name.replace("model.mm_projector", "mm")
name = re.sub(r'mm\.mlp\.mlp', 'mm.model.mlp', name, count=1)
name = re.sub(r'mm\.peg\.peg', 'mm.model.peg', name, count=1)
return name
return name.replace("text_model", "t").replace("vision_model", "v").replace("encoder.layers", "blk").replace("embeddings.", "").replace("_proj", "").replace("self_attn.", "attn_").replace("layer_norm", "ln").replace("layernorm", "ln").replace("mlp.fc1", "ffn_down").replace("mlp.fc2", "ffn_up").replace("embedding", "embd").replace("final", "post").replace("layrnorm", "ln")
def bytes_to_unicode():
"""
Returns list of utf-8 byte and a corresponding list of unicode strings.
The reversible bpe codes work on unicode strings.
This means you need a large # of unicode characters in your vocab if you want to avoid UNKs.
When you're at something like a 10B token dataset you end up needing around 5K for decent coverage.
This is a significant percentage of your normal, say, 32K bpe vocab.
To avoid that, we want lookup tables between utf-8 bytes and unicode strings.
And avoids mapping to whitespace/control characters the bpe code barfs on.
"""
bs = (
list(range(ord("!"), ord("~") + 1))
+ list(range(ord("¡"), ord("¬") + 1))
+ list(range(ord("®"), ord("ÿ") + 1))
)
cs = bs[:]
n = 0
for b in range(2**8):
if b not in bs:
bs.append(b)
cs.append(2**8 + n)
n += 1
cs = [chr(n) for n in cs]
return dict(zip(bs, cs))
ap = argparse.ArgumentParser()
ap.add_argument("-m", "--model-dir", help="Path to model directory cloned from HF Hub", required=True)
ap.add_argument("--use-f32", action="store_true", default=False, help="Use f32 instead of f16")
ap.add_argument("--text-only", action="store_true", required=False,
help="Save a text-only model. It can't be used to encode images")
ap.add_argument("--vision-only", action="store_true", required=False,
help="Save a vision-only model. It can't be used to encode texts")
ap.add_argument("--clip-model-is-vision", action="store_true", required=False,
help="The clip model is a pure vision model (ShareGPT4V vision extract for example)")
ap.add_argument("--clip-model-is-openclip", action="store_true", required=False,
help="The clip model is from openclip (for ViT-SO400M type))")
ap.add_argument("--llava-projector", help="Path to llava.projector file. If specified, save an image encoder for LLaVA models.")
ap.add_argument("--projector-type", help="Type of projector. Possible values: mlp, ldp, ldpv2", choices=["mlp", "ldp", "ldpv2","adapter"], default="adapter")
ap.add_argument("-o", "--output-dir", help="Directory to save GGUF files. Default is the original model directory", default=None)
# Example --image_mean 0.48145466 0.4578275 0.40821073 --image_std 0.26862954 0.26130258 0.27577711
# Example --image_mean 0.5 0.5 0.5 --image_std 0.5 0.5 0.5
default_image_mean = [0.5, 0.5, 0.5]
default_image_std = [0.5, 0.5, 0.5]
ap.add_argument('--image-mean', type=float, nargs='+', help='Mean of the images for normalization (overrides processor) ', default=None)
ap.add_argument('--image-std', type=float, nargs='+', help='Standard deviation of the images for normalization (overrides processor)', default=None)
# with proper
args = ap.parse_args()
if args.text_only and args.vision_only:
print("--text-only and --image-only arguments cannot be specified at the same time.")
exit(1)
if args.use_f32:
print("WARNING: Weights for the convolution op is always saved in f16, as the convolution op in GGML does not support 32-bit kernel weights yet.")
# output in the same directory as the model if output_dir is None
dir_model = args.model_dir
if args.clip_model_is_vision or not os.path.exists(dir_model + "/vocab.json") or args.clip_model_is_openclip:
vocab = None
tokens = None
else:
with open(dir_model + "/vocab.json", "r", encoding="utf-8") as f:
vocab = json.load(f)
tokens = [key for key in vocab]
with open(dir_model + "/config.json", "r", encoding="utf-8") as f:
config = json.load(f)
if args.clip_model_is_vision:
v_hparams = config
t_hparams = None
else:
v_hparams = config["vision_config"]
t_hparams = None
# possible data types
# ftype == 0 -> float32
# ftype == 1 -> float16
#
# map from ftype to string
ftype_str = ["f32", "f16"]
ftype = 1
if args.use_f32:
ftype = 0
vision_config = SiglipVisionConfig(**v_hparams)
model = SiglipVisionModel(vision_config)
model.load_state_dict(torch.load(os.path.join(dir_model, "glm.clip")))
fname_middle = None
has_text_encoder = False
has_vision_encoder = True
has_glm_projector = True
if args.text_only:
fname_middle = "text-"
has_vision_encoder = False
elif args.llava_projector is not None:
fname_middle = "mmproj-"
has_text_encoder = False
has_glm_projector = True
elif args.vision_only:
fname_middle = "vision-"
has_text_encoder = False
else:
fname_middle = ""
output_dir = args.output_dir if args.output_dir is not None else dir_model
os.makedirs(output_dir, exist_ok=True)
output_prefix = os.path.basename(output_dir).replace("ggml_", "")
fname_out = os.path.join(output_dir, f"{fname_middle}model-{ftype_str[ftype]}.gguf")
fout = GGUFWriter(path=fname_out, arch="clip")
fout.add_bool("clip.has_text_encoder", has_text_encoder)
fout.add_bool("clip.has_vision_encoder", has_vision_encoder)
fout.add_bool("clip.has_glm_projector", has_glm_projector)
fout.add_file_type(ftype)
model_name = config["_name_or_path"] if "_name_or_path" in config else os.path.basename(dir_model)
fout.add_name(model_name)
if has_glm_projector:
fout.add_description("image encoder for glm4v")
fout.add_string("clip.projector_type", "adapter")
else:
fout.add_description("two-tower CLIP model")
if has_text_encoder:
assert t_hparams is not None
assert tokens is not None
# text_model hparams
fout.add_uint32(k(KEY_CONTEXT_LENGTH, TEXT), t_hparams["max_position_embeddings"])
fout.add_uint32(k(KEY_EMBEDDING_LENGTH, TEXT), t_hparams["hidden_size"])
fout.add_uint32(k(KEY_FEED_FORWARD_LENGTH, TEXT), t_hparams["intermediate_size"])
fout.add_uint32("clip.text.projection_dim", t_hparams.get("projection_dim", config["projection_dim"]))
fout.add_uint32(k(KEY_ATTENTION_HEAD_COUNT, TEXT), t_hparams["num_attention_heads"])
fout.add_float32(k(KEY_ATTENTION_LAYERNORM_EPS, TEXT), t_hparams["layer_norm_eps"])
fout.add_uint32(k(KEY_BLOCK_COUNT, TEXT), t_hparams["num_hidden_layers"])
fout.add_token_list(tokens)
if has_vision_encoder:
# vision_model hparams
fout.add_uint32("clip.vision.image_size", v_hparams["image_size"])
fout.add_uint32("clip.vision.patch_size", v_hparams["patch_size"])
fout.add_uint32(k(KEY_EMBEDDING_LENGTH, VISION), v_hparams["hidden_size"])
fout.add_uint32(k(KEY_FEED_FORWARD_LENGTH, VISION), v_hparams["intermediate_size"])
fout.add_uint32("clip.vision.projection_dim", 0)
fout.add_uint32(k(KEY_ATTENTION_HEAD_COUNT, VISION), v_hparams["num_attention_heads"])
fout.add_float32(k(KEY_ATTENTION_LAYERNORM_EPS, VISION), 1e-6)
fout.add_uint32(k(KEY_BLOCK_COUNT, VISION), v_hparams["num_hidden_layers"])
image_mean = args.image_mean if args.image_mean is not None else default_image_mean
image_std = args.image_std if args.image_std is not None else default_image_std
fout.add_array("clip.vision.image_mean", image_mean)
fout.add_array("clip.vision.image_std", image_std)
fout.add_bool("clip.use_gelu", True)
if has_glm_projector:
# model.vision_model.encoder.layers.pop(-1) # pyright: ignore[reportAttributeAccessIssue]
projector = torch.load(args.llava_projector)
for name, data in projector.items():
name = get_tensor_name(name)
# pw and dw conv ndim==4
if data.ndim == 2 or data.ndim == 4:
data = data.squeeze().numpy().astype(np.float16)
else:
data = data.squeeze().numpy().astype(np.float32)
if name.startswith("vision."):
name=name.replace("vision.","")
fout.add_tensor(name, data)
print(f"Projector {name} - {data.dtype} - shape = {data.shape}")
# print(f"Projector {name} tensors added\n")
state_dict = model.state_dict() # pyright: ignore[reportAttributeAccessIssue]
for name, data in state_dict.items():
if should_skip_tensor(name, has_text_encoder, has_vision_encoder, has_glm_projector):
# we don't need this
print(f"skipping parameter: {name}")
continue
name = get_tensor_name(name)
data = data.squeeze().numpy()
n_dims = len(data.shape)
# ftype == 0 -> float32, ftype == 1 -> float16
ftype_cur = 0
if n_dims == 4:
print(f"tensor {name} is always saved in f16")
data = data.astype(np.float16)
ftype_cur = 1
elif ftype == 1:
if name[-7:] == ".weight" and n_dims == 2:
# print(" Converting to float16")
data = data.astype(np.float16)
ftype_cur = 1
else:
# print(" Converting to float32")
data = data.astype(np.float32)
ftype_cur = 0
else:
if data.dtype != np.float32:
# print(" Converting to float32")
data = data.astype(np.float32)
ftype_cur = 0
print(f"siglip {name} - {data.dtype} - shape = {data.shape}")
# print(f"{name} - {ftype_str[ftype_cur]} - shape = {data.shape}")
fout.add_tensor(name, data)
fout.write_header_to_file()
fout.write_kv_data_to_file()
fout.write_tensors_to_file()
fout.close()
print("Done. Output file: " + fname_out)

33
examples/llava/glmedge-surgery.py Normal file
View file

@ -0,0 +1,33 @@
import argparse
import os
import torch
from transformers import AutoModel
ap = argparse.ArgumentParser()
ap.add_argument("-m", "--model", help="Path to GLM model")
args = ap.parse_args()
# find the model part that includes the the multimodal projector weights
model = AutoModel.from_pretrained(args.model, trust_remote_code=True, local_files_only=True)
checkpoint = model.state_dict()
# get a list of mm tensor names
mm_tensors = [k for k, v in checkpoint.items() if k.startswith("vision.adapter.")]
# store these tensors in a new dictionary and torch.save them
projector = {name: checkpoint[name].float() for name in mm_tensors}
torch.save(projector, f"{args.model}/glm.projector")
clip_tensors = [k for k, v in checkpoint.items() if k.startswith("vision.vit.model.vision_model.")]
if len(clip_tensors) > 0:
clip = {name.replace("vision.vit.model.", ""): checkpoint[name].float() for name in clip_tensors}
torch.save(clip, f"{args.model}/glm.clip")
# added tokens should be removed to be able to convert Mistral models
if os.path.exists(f"{args.model}/added_tokens.json"):
with open(f"{args.model}/added_tokens.json", "w") as f:
f.write("{}\n")
print("Done!")
print(f"Now you can convert {args.model} to a regular LLaMA GGUF file.")
print(f"Also, use {args.model}glm.projector to prepare a glm-encoder.gguf file.")

17
examples/llava/llava.cpp
View file

@ -311,6 +311,20 @@ static bool encode_image_with_clip(clip_ctx * ctx_clip, int n_threads, const cli
img_res_v.size = 0;
img_res_v.data = nullptr;
}
else if (clip_is_glm(ctx_clip)){
struct clip_image_size * load_image_size = clip_image_size_init();
load_image_size->width = img_res_v.data[0].nx;
load_image_size->height = img_res_v.data[0].ny;
clip_add_load_image_size(ctx_clip, load_image_size);
bool encoded = clip_image_encode(ctx_clip, n_threads, &img_res_v.data[0], image_embd);
int pos = int(load_image_size->width/clip_patch_size(ctx_clip)/2);
*n_img_pos = (pos * pos + 2);
if (!encoded){
LOG_ERR("Unable to encode image \n");
return false;
}
}
else if (strcmp(mm_patch_merge_type, "spatial_unpad") != 0) {
// flat / default llava-1.5 type embedding
*n_img_pos = clip_n_patches(ctx_clip);
@ -395,6 +409,9 @@ bool llava_image_embed_make_with_clip_img(clip_ctx * ctx_clip, int n_threads, co
if (clip_is_minicpmv(ctx_clip)) {
num_max_patches = 10;
}
if (clip_is_glm(ctx_clip)) {
num_max_patches = 1;
}
float * image_embd;
if (clip_is_qwen2vl(ctx_clip)) {
// qwen2vl don't split image into chunks, so `num_max_patches` is not needed.

26
examples/run/linenoise.cpp/LICENSE
View file

@ -1,26 +0,0 @@
Copyright (c) 2010-2014, Salvatore Sanfilippo <antirez at gmail dot com>
Copyright (c) 2010-2013, Pieter Noordhuis <pcnoordhuis at gmail dot com>
Copyright (c) 2025, Eric Curtin <ericcurtin17 at gmail dot com>
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

1350
examples/run/linenoise.cpp/linenoise.cpp
View file

File diff suppressed because it is too large Load diff

128
examples/run/linenoise.cpp/linenoise.h
View file

@ -1,128 +0,0 @@
/* linenoise.h -- VERSION 1.0
*
* Guerrilla line editing library against the idea that a line editing lib
* needs to be 20,000 lines of C++ code.
*
* See linenoise.cpp for more information.
*
* ------------------------------------------------------------------------
*
* Copyright (c) 2010-2023, Salvatore Sanfilippo <antirez at gmail dot com>
* Copyright (c) 2010-2013, Pieter Noordhuis <pcnoordhuis at gmail dot com>
* Copyright (c) 2025, Eric Curtin <ericcurtin17 at gmail dot com>
*
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are
* met:
*
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* HOLDER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef __LINENOISE_H
#define __LINENOISE_H
#ifdef __cplusplus
extern "C" {
#endif
#include <stddef.h> /* For size_t. */
#include <stdlib.h>
extern const char *linenoiseEditMore;
/* The linenoiseState structure represents the state during line editing.
* We pass this state to functions implementing specific editing
* functionalities. */
struct linenoiseState {
int in_completion; /* The user pressed TAB and we are now in completion
* mode, so input is handled by completeLine(). */
size_t completion_idx; /* Index of next completion to propose. */
int ifd; /* Terminal stdin file descriptor. */
int ofd; /* Terminal stdout file descriptor. */
char *buf; /* Edited line buffer. */
size_t buflen; /* Edited line buffer size. */
const char *prompt; /* Prompt to display. */
size_t plen; /* Prompt length. */
size_t pos; /* Current cursor position. */
size_t oldpos; /* Previous refresh cursor position. */
size_t len; /* Current edited line length. */
size_t cols; /* Number of columns in terminal. */
size_t oldrows; /* Rows used by last refrehsed line (multiline mode) */
int history_index; /* The history index we are currently editing. */
};
struct linenoiseCompletions {
size_t len = 0;
char ** cvec = nullptr;
bool to_free = true;
~linenoiseCompletions() {
if (!to_free) {
return;
}
for (size_t i = 0; i < len; ++i) {
free(cvec[i]);
}
free(cvec);
}
};
/* Non blocking API. */
int linenoiseEditStart(struct linenoiseState *l, int stdin_fd, int stdout_fd, char *buf, size_t buflen, const char *prompt);
const char *linenoiseEditFeed(struct linenoiseState *l);
void linenoiseEditStop(struct linenoiseState *l);
void linenoiseHide(struct linenoiseState *l);
void linenoiseShow(struct linenoiseState *l);
/* Blocking API. */
const char *linenoise(const char *prompt);
void linenoiseFree(void *ptr);
/* Completion API. */
typedef void(linenoiseCompletionCallback)(const char *, linenoiseCompletions *);
typedef const char*(linenoiseHintsCallback)(const char *, int *color, int *bold);
typedef void(linenoiseFreeHintsCallback)(const char *);
void linenoiseSetCompletionCallback(linenoiseCompletionCallback *);
void linenoiseSetHintsCallback(linenoiseHintsCallback *);
void linenoiseSetFreeHintsCallback(linenoiseFreeHintsCallback *);
void linenoiseAddCompletion(linenoiseCompletions *, const char *);
/* History API. */
int linenoiseHistoryAdd(const char *line);
int linenoiseHistorySetMaxLen(int len);
int linenoiseHistorySave(const char *filename);
int linenoiseHistoryLoad(const char *filename);
/* Other utilities. */
void linenoiseClearScreen(void);
void linenoiseSetMultiLine(int ml);
void linenoisePrintKeyCodes(void);
void linenoiseMaskModeEnable(void);
void linenoiseMaskModeDisable(void);
#ifdef __cplusplus
}
#endif
#endif /* __LINENOISE_H */

BIN
examples/server/public/index.html.gz
View file

Binary file not shown.

58
examples/server/server.cpp
View file

@ -131,6 +131,11 @@ struct slot_params {
lora.push_back({{"id", i}, {"scale", this->lora[i].scale}});
}
std::vector<std::string> grammar_trigger_words;
for (const auto & trigger : sampling.grammar_trigger_words) {
grammar_trigger_words.push_back(trigger.word);
}
return json {
{"n_predict", n_predict}, // Server configured n_predict
{"seed", sampling.seed},
@ -165,8 +170,9 @@ struct slot_params {
{"n_probs", sampling.n_probs},
{"min_keep", sampling.min_keep},
{"grammar", sampling.grammar},
// {"grammar_trigger_words", sampling.grammar_trigger_words},
{"grammar_trigger_words", grammar_trigger_words},
{"grammar_trigger_tokens", sampling.grammar_trigger_tokens},
{"preserved_tokens", sampling.preserved_tokens},
{"samplers", samplers},
{"speculative.n_max", speculative.n_max},
{"speculative.n_min", speculative.n_min},
@ -363,12 +369,26 @@ struct server_task {
if (ids.size() == 1) {
LOG_DBG("Grammar trigger token: %d (`%s`)\n", ids[0], trigger.word.c_str());
params.sampling.grammar_trigger_tokens.push_back(ids[0]);
params.sampling.preserved_tokens.insert(ids[0]);
continue;
}
LOG_DBG("Grammar trigger word: `%s`\n", trigger.word.c_str());
params.sampling.grammar_trigger_words.push_back(trigger);
}
}
const auto preserved_tokens = data.find("preserved_tokens");
if (preserved_tokens != data.end()) {
for (const auto & t : *preserved_tokens) {
auto ids = common_tokenize(vocab, t.get<std::string>(), /* add_special= */ false, /* parse_special= */ true);
if (ids.size() == 1) {
LOG_DBG("Preserved token: %d\n", ids[0]);
params.sampling.preserved_tokens.insert(ids[0]);
} else {
// This may happen when using a tool call style meant for a model with special tokens to preserve on a model without said tokens.
LOG_WRN("Not preserved because more than 1 token (wrong chat template override?): %s\n", t.get<std::string>().c_str());
}
}
}
if (params.sampling.grammar_lazy) {
GGML_ASSERT(params.sampling.grammar_trigger_tokens.size() > 0 || params.sampling.grammar_trigger_words.size() > 0);
}
@ -695,19 +715,19 @@ struct server_task_result_cmpl_final : server_task_result {
json to_json_oaicompat_chat() {
std::string finish_reason = "length";
common_chat_msg message;
common_chat_msg msg;
if (stop == STOP_TYPE_WORD || stop == STOP_TYPE_EOS) {
LOG_DBG("Parsing chat message: %s\n", content.c_str());
message = common_chat_parse(content, oaicompat_chat_format);
finish_reason = message.tool_calls.empty() ? "stop" : "tool_calls";
msg = common_chat_parse(content, oaicompat_chat_format);
finish_reason = msg.tool_calls.empty() ? "stop" : "tool_calls";
} else {
message.content = content;
msg.content = content;
}
json tool_calls;
if (!message.tool_calls.empty()) {
if (!msg.tool_calls.empty()) {
tool_calls = json::array();
for (const auto & tc : message.tool_calls) {
for (const auto & tc : msg.tool_calls) {
tool_calls.push_back({
{"type", "function"},
{"function", {
@ -719,14 +739,19 @@ struct server_task_result_cmpl_final : server_task_result {
}
}
json message {
{"content", msg.content},
{"tool_calls", tool_calls},
{"role", "assistant"},
};
if (!msg.tool_plan.empty()) {
message["tool_plan"] = msg.tool_plan;
}
json choice {
{"finish_reason", finish_reason},
{"index", 0},
{"message", json {
{"content", message.content},
{"tool_calls", tool_calls},
{"role", "assistant"},
}},
{"message", message},
};
if (!stream && probs_output.size() > 0) {
@ -2833,8 +2858,7 @@ struct server_context {
server_slot * slot_batched = nullptr;
auto accept_special_token = [&](server_slot & slot, llama_token token) {
const auto & trigger_tokens = slot.params.sampling.grammar_trigger_tokens;
return params_base.special || std::find(trigger_tokens.begin(), trigger_tokens.end(), token) != trigger_tokens.end();
return params_base.special || slot.params.sampling.preserved_tokens.find(token) != slot.params.sampling.preserved_tokens.end();
};
// frist, add sampled tokens from any ongoing sequences
@ -3329,6 +3353,8 @@ static void log_server_request(const httplib::Request & req, const httplib::Resp
return;
}
// reminder: this function is not covered by httplib's exception handler; if someone does more complicated stuff, think about wrapping it in try-catch
LOG_INF("request: %s %s %s %d\n", req.method.c_str(), req.path.c_str(), req.remote_addr.c_str(), res.status);
LOG_DBG("request: %s\n", req.body.c_str());
@ -3415,9 +3441,13 @@ int main(int argc, char ** argv) {
message = "Unknown Exception";
}
try {
json formatted_error = format_error_response(message, ERROR_TYPE_SERVER);
LOG_WRN("got exception: %s\n", formatted_error.dump().c_str());
res_error(res, formatted_error);
} catch (const std::exception & e) {
LOG_ERR("got another exception: %s | while hanlding exception: %s\n", e.what(), message.c_str());
}
});
svr->set_error_handler([&res_error](const httplib::Request &, httplib::Response & res) {

3
examples/server/tests/unit/test_chat_completion.py
View file

@ -13,8 +13,11 @@ def create_server():
@pytest.mark.parametrize(
"model,system_prompt,user_prompt,max_tokens,re_content,n_prompt,n_predicted,finish_reason,jinja,chat_template",
[
(None, "Book", "Hey", 8, "But she couldn't", 69, 8, "length", False, None),
(None, "Book", "Hey", 8, "But she couldn't", 69, 8, "length", True, None),
(None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", False, None),
(None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", True, None),
(None, "Book", "What is the best book", 8, "(Suddenly)+|\\{ \" Sarax.", 77, 8, "length", True, 'chatml'),
(None, "Book", "What is the best book", 8, "^ blue", 23, 8, "length", True, "This is not a chat template, it is"),
("codellama70b", "You are a coding assistant.", "Write the fibonacci function in c++.", 128, "(Aside|she|felter|alonger)+", 104, 64, "length", False, None),
("codellama70b", "You are a coding assistant.", "Write the fibonacci function in c++.", 128, "(Aside|she|felter|alonger)+", 104, 64, "length", True, None),

96
examples/server/tests/unit/test_tool_call.py
View file

@ -67,8 +67,8 @@ WEATHER_TOOL = {
def do_test_completion_with_required_tool_tiny(template_name: str, tool: dict, argument_key: str | None):
n_predict = 512
global server
n_predict = 512
# server = ServerPreset.stories15m_moe()
server.jinja = True
server.n_predict = n_predict
@ -139,29 +139,49 @@ def test_completion_with_required_tool_tiny_slow(template_name: str, tool: dict,
@pytest.mark.parametrize("tool,argument_key,hf_repo,template_override", [
(TEST_TOOL, "success", "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
# Note: gemma-2-2b-it knows itself as "model", not "assistant", so we don't test the ill-suited chatml on it.
(TEST_TOOL, "success", "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
(TEST_TOOL, "success", "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
(PYTHON_TOOL, "code", "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
(PYTHON_TOOL, "code", "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-3-Llama-3.1-8B", "tool_use")),
(PYTHON_TOOL, "code", "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-3-Llama-3.1-8B", "tool_use")),
(PYTHON_TOOL, "code", "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
(PYTHON_TOOL, "code", "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
(TEST_TOOL, "success", "bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai/functionary-medium-v3.2", None)),
(PYTHON_TOOL, "code", "bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai/functionary-medium-v3.2", None)),
(PYTHON_TOOL, "code", "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/functionary-small-v3.2-GGUF:Q4_K_M", ("meetkai/functionary-medium-v3.2", None)),
(PYTHON_TOOL, "code", "bartowski/functionary-small-v3.2-GGUF:Q4_K_M", ("meetkai/functionary-medium-v3.2", None)),
(PYTHON_TOOL, "code", "bartowski/functionary-small-v3.2-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
(PYTHON_TOOL, "code", "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
(PYTHON_TOOL, "code", "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
(TEST_TOOL, "success", "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
(PYTHON_TOOL, "code", "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
(PYTHON_TOOL, "code", "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", "chatml"),
# TODO: fix these
# (TEST_TOOL, "success", "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
# (PYTHON_TOOL, "code", "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
])
def test_completion_with_required_tool_real_model(tool: dict, argument_key: str | None, hf_repo: str, template_override: Tuple[str, str | None] | None):
def test_completion_with_required_tool_real_model(tool: dict, argument_key: str | None, hf_repo: str, template_override: str | Tuple[str, str | None] | None):
global server
n_predict = 512
server.n_slots = 1
server.jinja = True
@ -169,10 +189,12 @@ def test_completion_with_required_tool_real_model(tool: dict, argument_key: str
server.n_predict = n_predict
server.model_hf_repo = hf_repo
server.model_hf_file = None
if template_override:
if isinstance(template_override, tuple):
(template_hf_repo, template_variant) = template_override
server.chat_template_file = f"../../../models/templates/{template_hf_repo.replace('/', '-') + ('-' + template_variant if template_variant else '')}.jinja"
assert os.path.exists(server.chat_template_file), f"Template file {server.chat_template_file} does not exist. Run `python scripts/get_chat_template.py {template_hf_repo} {template_variant} > {server.chat_template_file}` to download the template."
elif isinstance(template_override, str):
server.chat_template = template_override
server.start(timeout_seconds=TIMEOUT_SERVER_START)
res = server.make_request("POST", "/chat/completions", data={
"max_tokens": n_predict,
@ -251,33 +273,55 @@ def test_completion_without_tool_call_slow(template_name: str, n_predict: int, t
@pytest.mark.slow
@pytest.mark.parametrize("hf_repo,template_override", [
("bartowski/c4ai-command-r7b-12-2024-GGUF:Q4_K_M", ("CohereForAI/c4ai-command-r7b-12-2024", "tool_use")),
("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
("bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
("bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
("bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
("bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
("bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
("bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
("bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
("bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", "chatml"),
("bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-3-Llama-3.1-8B", "tool_use")),
("bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", "chatml"),
("bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
("bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", "chatml"),
("bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai/functionary-medium-v3.2", None)),
("bartowski/functionary-small-v3.2-GGUF:Q8_0", "chatml"),
("bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
("bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
# Note: gemma-2-2b-it knows itself as "model", not "assistant", so we don't test the ill-suited chatml on it.
("bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
# ("bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama/Llama-3.2-3B-Instruct", None)),
# ("bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
])
def test_weather_tool_call(hf_repo: str, template_override: Tuple[str, str | None] | None):
def test_weather(hf_repo: str, template_override: Tuple[str, str | None] | None):
global server
n_predict = 512
server.n_slots = 1
server.jinja = True
server.n_ctx = 8192
server.n_predict = 512
server.n_predict = n_predict
server.model_hf_repo = hf_repo
server.model_hf_file = None
if template_override:
if isinstance(template_override, tuple):
(template_hf_repo, template_variant) = template_override
server.chat_template_file = f"../../../models/templates/{template_hf_repo.replace('/', '-') + ('-' + template_variant if template_variant else '')}.jinja"
assert os.path.exists(server.chat_template_file), f"Template file {server.chat_template_file} does not exist. Run `python scripts/get_chat_template.py {template_hf_repo} {template_variant} > {server.chat_template_file}` to download the template."
elif isinstance(template_override, str):
server.chat_template = template_override
server.start(timeout_seconds=TIMEOUT_SERVER_START)
res = server.make_request("POST", "/chat/completions", data={
"max_tokens": 256,
"max_tokens": n_predict,
"messages": [
{"role": "user", "content": "What is the weather in Istanbul?"},
],
@ -298,19 +342,39 @@ def test_weather_tool_call(hf_repo: str, template_override: Tuple[str, str | Non
@pytest.mark.slow
@pytest.mark.parametrize("expected_arguments_override,hf_repo,template_override", [
(None, "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", None),
(None, "bartowski/Phi-3.5-mini-instruct-GGUF:Q4_K_M", "chatml"),
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", ("meetkai-functionary-medium-v3.2", None)),
('{"code":"print("}', "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
(None, "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
(None, "bartowski/functionary-small-v3.2-GGUF:Q8_0", "chatml"),
(None, "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", None),
('{"code":"print("}', "bartowski/Meta-Llama-3.1-8B-Instruct-GGUF:Q4_K_M", "chatml"),
('{"code":"print("}', "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
(None, "bartowski/Llama-3.2-1B-Instruct-GGUF:Q4_K_M", "chatml"),
('{"code":"print("}', "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", ("meta-llama-Llama-3.2-3B-Instruct", None)),
('{"code":"print("}', "bartowski/Llama-3.2-3B-Instruct-GGUF:Q4_K_M", "chatml"),
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", None),
(None, "bartowski/Qwen2.5-7B-Instruct-GGUF:Q4_K_M", "chatml"),
(None, "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", ("NousResearch/Hermes-2-Pro-Llama-3-8B", "tool_use")),
(None, "bartowski/Hermes-2-Pro-Llama-3-8B-GGUF:Q4_K_M", "chatml"),
(None, "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", ("NousResearch-Hermes-3-Llama-3.1-8B", "tool_use")),
(None, "bartowski/Hermes-3-Llama-3.1-8B-GGUF:Q4_K_M", "chatml"),
(None, "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", None),
(None, "bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M", "chatml"),
# Note: gemma-2-2b-it knows itself as "model", not "assistant", so we don't test the ill-suited chatml on it.
(None, "bartowski/gemma-2-2b-it-GGUF:Q4_K_M", None),
# (None, "bartowski/DeepSeek-R1-Distill-Qwen-7B-GGUF:Q4_K_M", None),
])
def test_hello_world_tool_call(expected_arguments_override: str | None, hf_repo: str, template_override: Tuple[str, str | None] | None):
def test_hello_world_tool_call(expected_arguments_override: str | None, hf_repo: str, template_override: str | Tuple[str, str | None] | None):
global server
server.n_slots = 1
server.jinja = True
@ -318,10 +382,12 @@ def test_hello_world_tool_call(expected_arguments_override: str | None, hf_repo:
server.n_predict = 128
server.model_hf_repo = hf_repo
server.model_hf_file = None
if template_override:
if isinstance(template_override, tuple):
(template_hf_repo, template_variant) = template_override
server.chat_template_file = f"../../../models/templates/{template_hf_repo.replace('/', '-') + ('-' + template_variant if template_variant else '')}.jinja"
assert os.path.exists(server.chat_template_file), f"Template file {server.chat_template_file} does not exist. Run `python scripts/get_chat_template.py {template_hf_repo} {template_variant} > {server.chat_template_file}` to download the template."
elif isinstance(template_override, str):
server.chat_template = template_override
server.start(timeout_seconds=TIMEOUT_SERVER_START)
res = server.make_request("POST", "/chat/completions", data={
"max_tokens": 256,

5
examples/server/utils.hpp
View file

@ -5,10 +5,6 @@
#include "llama.h"
#include "common/base64.hpp"
#ifndef NDEBUG
// crash the server in debug mode, otherwise send an http 500 error
#define CPPHTTPLIB_NO_EXCEPTIONS 1
#endif
// increase max payload length to allow use of larger context size
#define CPPHTTPLIB_FORM_URL_ENCODED_PAYLOAD_MAX_LENGTH 1048576
#include "httplib.h"
@ -662,6 +658,7 @@ static json oaicompat_completion_params_parse(
});
}
llama_params["grammar_triggers"] = grammar_triggers;
llama_params["preserved_tokens"] = chat_params.preserved_tokens;
for (const auto & stop : chat_params.additional_stops) {
llama_params["stop"].push_back(stop);
}

24
examples/server/webui/.gitignore vendored Normal file
View file

@ -0,0 +1,24 @@
# Logs
logs
*.log
npm-debug.log*
yarn-debug.log*
yarn-error.log*
pnpm-debug.log*
lerna-debug.log*
node_modules
dist
dist-ssr
*.local
# Editor directories and files
.vscode/*
!.vscode/extensions.json
.idea
.DS_Store
*.suo
*.ntvs*
*.njsproj
*.sln
*.sw?

10
examples/server/webui/.prettierignore Normal file
View file

@ -0,0 +1,10 @@
**/.vscode
**/.github
**/.git
**/.svn
**/.hg
**/node_modules
**/dist
**/build
*.config.js

26
examples/server/webui/eslint.config.js Normal file
View file

@ -0,0 +1,26 @@
import js from '@eslint/js'
import globals from 'globals'
import reactHooks from 'eslint-plugin-react-hooks'
import reactRefresh from 'eslint-plugin-react-refresh'
import tseslint from 'typescript-eslint'
export default tseslint.config(
{ ignores: ['dist'] },
{
extends: [js.configs.recommended, ...tseslint.configs.recommended],
files: ['**/*.{ts,tsx}'],
languageOptions: {
ecmaVersion: 2020,
globals: globals.browser,
},
plugins: {
'react-hooks': reactHooks,
'react-refresh': reactRefresh,
},
rules: {
...reactHooks.configs.recommended.rules,
'react-refresh/only-export-components': 'off',
'@typescript-eslint/no-unused-vars': 'off',
},
},
)

345
examples/server/webui/index.html
View file

@ -1,343 +1,16 @@
<!DOCTYPE html>
<!doctype html>
<html>
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1, maximum-scale=1" />
<meta name="color-scheme" content="light dark">
<meta charset="UTF-8" />
<meta
name="viewport"
content="width=device-width, initial-scale=1, maximum-scale=1"
/>
<meta name="color-scheme" content="light dark" />
<title>🦙 llama.cpp - chat</title>
</head>
<body>
<div id="app" class="opacity-0"> <!-- opacity-0 will be removed on app mounted -->
<div class="flex flex-row drawer lg:drawer-open">
<input id="toggle-drawer" type="checkbox" class="drawer-toggle" checked />
<!-- sidebar -->
<div class="drawer-side h-screen lg:h-screen z-50 lg:max-w-64">
<label for="toggle-drawer" aria-label="close sidebar" class="drawer-overlay"></label>
<div class="flex flex-col bg-base-200 min-h-full max-w-64 py-4 px-4">
<div class="flex flex-row items-center justify-between mb-4 mt-4">
<h2 class="font-bold ml-4">Conversations</h2>
<!-- close sidebar button -->
<label for="toggle-drawer" class="btn btn-ghost lg:hidden">
<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" fill="currentColor" class="bi bi-arrow-bar-left" viewBox="0 0 16 16">
<path fill-rule="evenodd" d="M12.5 15a.5.5 0 0 1-.5-.5v-13a.5.5 0 0 1 1 0v13a.5.5 0 0 1-.5.5M10 8a.5.5 0 0 1-.5.5H3.707l2.147 2.146a.5.5 0 0 1-.708.708l-3-3a.5.5 0 0 1 0-.708l3-3a.5.5 0 1 1 .708.708L3.707 7.5H9.5a.5.5 0 0 1 .5.5"/>
</svg>
</label>
</div>
<!-- list of conversations -->
<div :class="{
'btn btn-ghost justify-start': true,
'btn-active': messages.length === 0,
}" @click="newConversation">
+ New conversation
</div>
<div v-for="conv in conversations" :class="{
'btn btn-ghost justify-start font-normal': true,
'btn-active': conv.id === viewingConvId,
}" @click="setViewingConv(conv.id)" dir="auto">
<span class="truncate">{{ conv.messages[0].content }}</span>
</div>
<div class="text-center text-xs opacity-40 mt-auto mx-4">
Conversations are saved to browser's localStorage
</div>
</div>
</div>
<!-- main view -->
<div class="chat-screen drawer-content grow flex flex-col h-screen w-screen mx-auto px-4">
<!-- header -->
<div class="flex flex-row items-center mt-6 mb-6">
<!-- open sidebar button -->
<label for="toggle-drawer" class="btn btn-ghost lg:hidden">
<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" fill="currentColor" class="bi bi-list" viewBox="0 0 16 16">
<path fill-rule="evenodd" d="M2.5 12a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5m0-4a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5m0-4a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5"/>
</svg>
</label>
<div class="grow text-2xl font-bold ml-2">llama.cpp</div>
<!-- action buttons (top right) -->
<div class="flex items-center">
<div v-if="messages.length > 0" class="dropdown dropdown-end">
<!-- "..." button -->
<button tabindex="0" role="button" class="btn m-1" :disabled="isGenerating">
<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" fill="currentColor" class="bi bi-three-dots-vertical" viewBox="0 0 16 16">
<path d="M9.5 13a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0m0-5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0m0-5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0"/>
</svg>
</button>
<!-- "delete" dropdown menu -->
<ul tabindex="0" class="dropdown-content menu bg-base-100 rounded-box z-[1] w-52 p-2 shadow">
<li @click="downloadConv(viewingConvId)"><a>Download</a></li>
<li class="text-error" @click="deleteConv(viewingConvId)"><a>Delete</a></li>
</ul>
</div>
<div class="tooltip tooltip-bottom" data-tip="Settings">
<button class="btn" @click="showConfigDialog = true" :disabled="isGenerating">
<!-- settings button -->
<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" fill="currentColor" class="bi bi-gear" viewBox="0 0 16 16">
<path d="M8 4.754a3.246 3.246 0 1 0 0 6.492 3.246 3.246 0 0 0 0-6.492M5.754 8a2.246 2.246 0 1 1 4.492 0 2.246 2.246 0 0 1-4.492 0"/>
<path d="M9.796 1.343c-.527-1.79-3.065-1.79-3.592 0l-.094.319a.873.873 0 0 1-1.255.52l-.292-.16c-1.64-.892-3.433.902-2.54 2.541l.159.292a.873.873 0 0 1-.52 1.255l-.319.094c-1.79.527-1.79 3.065 0 3.592l.319.094a.873.873 0 0 1 .52 1.255l-.16.292c-.892 1.64.901 3.434 2.541 2.54l.292-.159a.873.873 0 0 1 1.255.52l.094.319c.527 1.79 3.065 1.79 3.592 0l.094-.319a.873.873 0 0 1 1.255-.52l.292.16c1.64.893 3.434-.902 2.54-2.541l-.159-.292a.873.873 0 0 1 .52-1.255l.319-.094c1.79-.527 1.79-3.065 0-3.592l-.319-.094a.873.873 0 0 1-.52-1.255l.16-.292c.893-1.64-.902-3.433-2.541-2.54l-.292.159a.873.873 0 0 1-1.255-.52zm-2.633.283c.246-.835 1.428-.835 1.674 0l.094.319a1.873 1.873 0 0 0 2.693 1.115l.291-.16c.764-.415 1.6.42 1.184 1.185l-.159.292a1.873 1.873 0 0 0 1.116 2.692l.318.094c.835.246.835 1.428 0 1.674l-.319.094a1.873 1.873 0 0 0-1.115 2.693l.16.291c.415.764-.42 1.6-1.185 1.184l-.291-.159a1.873 1.873 0 0 0-2.693 1.116l-.094.318c-.246.835-1.428.835-1.674 0l-.094-.319a1.873 1.873 0 0 0-2.692-1.115l-.292.16c-.764.415-1.6-.42-1.184-1.185l.159-.291A1.873 1.873 0 0 0 1.945 8.93l-.319-.094c-.835-.246-.835-1.428 0-1.674l.319-.094A1.873 1.873 0 0 0 3.06 4.377l-.16-.292c-.415-.764.42-1.6 1.185-1.184l.292.159a1.873 1.873 0 0 0 2.692-1.115z"/>
</svg>
</button>
</div>
<!-- theme controller is copied from https://daisyui.com/components/theme-controller/ -->
<div class="tooltip tooltip-bottom" data-tip="Themes">
<div class="dropdown dropdown-end dropdown-bottom">
<div tabindex="0" role="button" class="btn m-1">
<svg xmlns="http://www.w3.org/2000/svg" width="16" height="16" fill="currentColor" class="bi bi-palette2" viewBox="0 0 16 16">
<path d="M0 .5A.5.5 0 0 1 .5 0h5a.5.5 0 0 1 .5.5v5.277l4.147-4.131a.5.5 0 0 1 .707 0l3.535 3.536a.5.5 0 0 1 0 .708L10.261 10H15.5a.5.5 0 0 1 .5.5v5a.5.5 0 0 1-.5.5H3a3 3 0 0 1-2.121-.879A3 3 0 0 1 0 13.044m6-.21 7.328-7.3-2.829-2.828L6 7.188zM4.5 13a1.5 1.5 0 1 0-3 0 1.5 1.5 0 0 0 3 0M15 15v-4H9.258l-4.015 4zM0 .5v12.495zm0 12.495V13z"/>
</svg>
</div>
<ul tabindex="0" class="dropdown-content bg-base-300 rounded-box z-[1] w-52 p-2 shadow-2xl h-80 overflow-y-auto">
<li>
<button
class="btn btn-sm btn-block btn-ghost justify-start"
:class="{ 'btn-active': selectedTheme === 'auto' }"
@click="setSelectedTheme('auto')">
auto
</button>
</li>
<li v-for="theme in themes">
<input
type="radio"
name="theme-dropdown"
class="theme-controller btn btn-sm btn-block btn-ghost justify-start"
:aria-label="theme"
:value="theme"
:checked="selectedTheme === theme"
@click="setSelectedTheme(theme)" />
</li>
</ul>
</div>
</div>
</div>
</div>
<!-- chat messages -->
<div id="messages-list" class="flex flex-col grow overflow-y-auto">
<div class="mt-auto flex justify-center">
<!-- placeholder to shift the message to the bottom -->
{{ messages.length === 0 ? 'Send a message to start' : '' }}
</div>
<div v-for="msg in messages" class="group">
<message-bubble
:config="config"
:msg="msg"
:key="msg.id"
:is-generating="isGenerating"
:edit-user-msg-and-regenerate="editUserMsgAndRegenerate"
:regenerate-msg="regenerateMsg"></message-bubble>
</div>
<!-- pending (ongoing) assistant message -->
<div id="pending-msg" class="group">
<message-bubble
v-if="pendingMsg"
:config="config"
:msg="pendingMsg"
:key="pendingMsg.id"
:is-generating="isGenerating"
:show-thought-in-progress="config.showThoughtInProgress"
:edit-user-msg-and-regenerate="() => {}"
:regenerate-msg="() => {}"></message-bubble>
</div>
</div>
<!-- chat input -->
<div class="flex flex-row items-center mt-8 mb-6">
<textarea
class="textarea textarea-bordered w-full"
placeholder="Type a message (Shift+Enter to add a new line)"
v-model="inputMsg"
@keydown.enter.exact.prevent="sendMessage"
@keydown.enter.shift.exact.prevent="inputMsg += '\n'"
:disabled="isGenerating"
id="msg-input"
dir="auto"
></textarea>
<button v-if="!isGenerating" class="btn btn-primary ml-2" @click="sendMessage" :disabled="inputMsg.length === 0">Send</button>
<button v-else class="btn btn-neutral ml-2" @click="stopGeneration">Stop</button>
</div>
</div>
</div>
<!-- modal for editing config -->
<dialog class="modal" :class="{'modal-open': showConfigDialog}">
<div class="modal-box">
<h3 class="text-lg font-bold mb-6">Settings</h3>
<div class="h-[calc(90vh-12rem)] overflow-y-auto">
<p class="opacity-40 mb-6">Settings below are saved in browser's localStorage</p>
<settings-modal-short-input :config-key="'apiKey'" :config-default="configDefault" :config-info="configInfo" v-model="config.apiKey"></settings-modal-short-input>
<label class="form-control mb-2">
<div class="label">System Message</div>
<textarea class="textarea textarea-bordered h-24" :placeholder="'Default: ' + configDefault.systemMessage" v-model="config.systemMessage"></textarea>
</label>
<template v-for="configKey in ['temperature', 'top_k', 'top_p', 'min_p', 'max_tokens']">
<settings-modal-short-input :config-key="configKey" :config-default="configDefault" :config-info="configInfo" v-model="config[configKey]"></settings-modal-short-input>
</template>
<!-- TODO: add more sampling-related configs, please regroup them into different "collapse" sections -->
<!-- Section: Other sampler settings -->
<details class="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary class="collapse-title font-bold">Other sampler settings</summary>
<div class="collapse-content">
<!-- Samplers queue -->
<settings-modal-short-input label="Samplers queue" :config-key="'samplers'" :config-default="configDefault" :config-info="configInfo" v-model="config.samplers"></settings-modal-short-input>
<!-- Samplers -->
<template v-for="configKey in ['dynatemp_range', 'dynatemp_exponent', 'typical_p', 'xtc_probability', 'xtc_threshold']">
<settings-modal-short-input :config-key="configKey" :config-default="configDefault" :config-info="configInfo" v-model="config[configKey]"></settings-modal-short-input>
</template>
</div>
</details>
<!-- Section: Penalties settings -->
<details class="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary class="collapse-title font-bold">Penalties settings</summary>
<div class="collapse-content">
<template v-for="configKey in ['repeat_last_n', 'repeat_penalty', 'presence_penalty', 'frequency_penalty', 'dry_multiplier', 'dry_base', 'dry_allowed_length', 'dry_penalty_last_n']">
<settings-modal-short-input :config-key="configKey" :config-default="configDefault" :config-info="configInfo" v-model="config[configKey]"></settings-modal-short-input>
</template>
</div>
</details>
<!-- Section: Reasoning models -->
<details class="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary class="collapse-title font-bold">Reasoning models</summary>
<div class="collapse-content">
<div class="flex flex-row items-center mb-2">
<input type="checkbox" class="checkbox" v-model="config.showThoughtInProgress" />
<span class="ml-4">Expand though process by default for generating message</span>
</div>
<div class="flex flex-row items-center mb-2">
<input type="checkbox" class="checkbox" v-model="config.excludeThoughtOnReq" />
<span class="ml-4">Exclude thought process when sending request to API (Recommended for DeepSeek-R1)</span>
</div>
</div>
</details>
<!-- Section: Advanced config -->
<details class="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary class="collapse-title font-bold">Advanced config</summary>
<div class="collapse-content">
<div class="flex flex-row items-center mb-2" v-if="isDev">
<!-- this button only shows in dev mode, used to import a demo conversation to test message rendering -->
<button class="btn" @click="debugImportDemoConv()">(debug) Import demo conversation</button>
</div>
<div class="flex flex-row items-center mb-2">
<input type="checkbox" class="checkbox" v-model="config.showTokensPerSecond" />
<span class="ml-4">Show tokens per second</span>
</div>
<label class="form-control mb-2">
<!-- Custom parameters input -->
<div class="label inline">Custom JSON config (For more info, refer to <a class="underline" href="https://github.com/ggerganov/llama.cpp/blob/master/examples/server/README.md" target="_blank" rel="noopener noreferrer">server documentation</a>)</div>
<textarea class="textarea textarea-bordered h-24" placeholder="Example: { &quot;mirostat&quot;: 1, &quot;min_p&quot;: 0.1 }" v-model="config.custom"></textarea>
</label>
</div>
</details>
</div>
<!-- action buttons -->
<div class="modal-action">
<button class="btn" @click="resetConfigDialog">Reset to default</button>
<button class="btn" @click="closeAndDiscardConfigDialog">Close</button>
<button class="btn btn-primary" @click="closeAndSaveConfigDialog">Save</button>
</div>
</div>
</dialog>
</div>
<!-- Template to be used as message bubble -->
<template id="message-bubble">
<div :class="{
'chat': true,
'chat-start': msg.role !== 'user',
'chat-end': msg.role === 'user',
}">
<div :class="{
'chat-bubble markdown': true,
'chat-bubble-base-300': msg.role !== 'user',
}">
<!-- textarea for editing message -->
<template v-if="editingContent !== null">
<textarea
dir="auto"
class="textarea textarea-bordered bg-base-100 text-base-content w-[calc(90vw-8em)] lg:w-96"
v-model="editingContent"></textarea>
<br/>
<button class="btn btn-ghost mt-2 mr-2" @click="editingContent = null">Cancel</button>
<button class="btn mt-2" @click="editMsg()">Submit</button>
</template>
<template v-else>
<!-- show loading dots for pending message -->
<span v-if="msg.content === null" class="loading loading-dots loading-md"></span>
<!-- render message as markdown -->
<div v-else dir="auto">
<details v-if="msg.role === 'assistant' && splitMsgContent.cot" class="collapse bg-base-200 collapse-arrow mb-4" :open="splitMsgContent.isThinking && showThoughtInProgress">
<summary class="collapse-title">
<span v-if="splitMsgContent.isThinking">
<span v-if="isGenerating" class="loading loading-spinner loading-md mr-2" style="vertical-align: middle;"></span>
<b>Thinking</b>
</span>
<b v-else>Thought Process</b>
</summary>
<vue-markdown :source="splitMsgContent.cot" dir="auto" class="collapse-content"></vue-markdown>
</details>
<vue-markdown :source="splitMsgContent.content"></vue-markdown>
</div>
<!-- render timings if enabled -->
<div class="dropdown dropdown-hover dropdown-top mt-2" v-if="timings && config.showTokensPerSecond">
<div tabindex="0" role="button" class="cursor-pointer font-semibold text-sm opacity-60">Speed: {{ timings.predicted_per_second.toFixed(1) }} t/s</div>
<div class="dropdown-content bg-base-100 z-10 w-64 p-2 shadow mt-4">
<b>Prompt</b><br/>
- Tokens: {{ timings.prompt_n }}<br/>
- Time: {{ timings.prompt_ms }} ms<br/>
- Speed: {{ timings.prompt_per_second.toFixed(1) }} t/s<br/>
<b>Generation</b><br/>
- Tokens: {{ timings.predicted_n }}<br/>
- Time: {{ timings.predicted_ms }} ms<br/>
- Speed: {{ timings.predicted_per_second.toFixed(1) }} t/s<br/>
</div>
</div>
</template>
</div>
</div>
<!-- actions for each message -->
<div :class="{'text-right': msg.role === 'user', 'opacity-0': isGenerating}" class="mx-4 mt-2 mb-2">
<!-- user message -->
<button v-if="msg.role === 'user'" class="badge btn-mini show-on-hover" @click="editingContent = msg.content" :disabled="isGenerating">
✍️ Edit
</button>
<!-- assistant message -->
<button v-if="msg.role === 'assistant'" class="badge btn-mini show-on-hover mr-2" @click="regenerateMsg(msg)" :disabled="isGenerating">
🔄 Regenerate
</button>
<button v-if="msg.role === 'assistant'" class="badge btn-mini show-on-hover mr-2" @click="copyMsg()" :disabled="isGenerating">
📋 Copy
</button>
</div>
</template>
<!-- Template to be used by settings modal -->
<template id="settings-modal-short-input">
<label class="input input-bordered join-item grow flex items-center gap-2 mb-2">
<!-- Show help message on hovering on the input label -->
<div class="dropdown dropdown-hover">
<div tabindex="0" role="button" class="font-bold">{{ label || configKey }}</div>
<div class="dropdown-content menu bg-base-100 rounded-box z-10 w-64 p-2 shadow mt-4">
{{ configInfo[configKey] || '(no help message available)' }}
</div>
</div>
<!-- Here we forward v-model from parent to child component, see: https://stackoverflow.com/questions/47311936/v-model-and-child-components -->
<input type="text" class="grow" :placeholder="'Default: ' + (configDefault[configKey] || 'none')" :value="modelValue" @input="$emit('update:modelValue', $event.target.value)" />
</label>
</template>
<script type="module" src="/src/main.js"></script>
<div id="root"></div>
<script type="module" src="/src/main.tsx"></script>
</body>
</html>

6649
examples/server/webui/package-lock.json generated
View file

File diff suppressed because it is too large Load diff

48
examples/server/webui/package.json
View file

@ -5,13 +5,10 @@
"type": "module",
"scripts": {
"dev": "vite",
"build": "vite build",
"preview": "vite preview",
"analyze": "ANALYZE=1 npx vite-bundle-visualizer"
},
"devDependencies": {
"sass-embedded": "^1.83.0",
"vite": "^5.4.10"
"build": "tsc -b && vite build",
"format": "eslint . && prettier --write .",
"lint": "eslint .",
"preview": "vite preview"
},
"dependencies": {
"@sec-ant/readable-stream": "^0.6.0",
@ -20,11 +17,42 @@
"daisyui": "^4.12.14",
"highlight.js": "^11.10.0",
"katex": "^0.16.15",
"markdown-it": "^14.1.0",
"postcss": "^8.4.49",
"react": "^18.3.1",
"react-dom": "^18.3.1",
"react-markdown": "^9.0.3",
"react-router": "^7.1.5",
"rehype-highlight": "^7.0.2",
"rehype-katex": "^7.0.1",
"remark-breaks": "^4.0.0",
"remark-gfm": "^4.0.0",
"remark-math": "^6.0.0",
"tailwindcss": "^3.4.15",
"textlinestream": "^1.1.1",
"vite-plugin-singlefile": "^2.0.3",
"vue": "^3.5.13"
"vite-plugin-singlefile": "^2.0.3"
},
"devDependencies": {
"@eslint/js": "^9.17.0",
"@types/markdown-it": "^14.1.2",
"@types/node": "^22.13.1",
"@types/react": "^18.3.18",
"@types/react-dom": "^18.3.5",
"@vitejs/plugin-react": "^4.3.4",
"eslint": "^9.17.0",
"eslint-plugin-react-hooks": "^5.0.0",
"eslint-plugin-react-refresh": "^0.4.16",
"globals": "^15.14.0",
"prettier": "^3.4.2",
"sass-embedded": "^1.83.4",
"typescript": "~5.6.2",
"typescript-eslint": "^8.18.2",
"vite": "^6.0.5"
},
"prettier": {
"trailingComma": "es5",
"tabWidth": 2,
"semi": true,
"singleQuote": true,
"bracketSameLine": false
}
}

2
examples/server/webui/public/demo-conversation.json
View file

@ -11,7 +11,7 @@
{
"id": 1734087548327,
"role": "assistant",
"content": "This is the formula:\n\n$\\frac{e^{x_i}}{\\sum_{j=1}^{n}e^{x_j}}$\n\nGiven an input vector \\(\\mathbf{x} = [x_1, x_2, \\ldots, x_n]\\)\n\n\\[\ny_i = \\frac{e^{x_i}}{\\sum_{j=1}^n e^{x_j}}\n\\]\n\nCode block latex:\n```latex\n\\frac{e^{x_i}}{\\sum_{j=1}^{n}e^{x_j}}\n```\n\nTest dollar sign: $1234 $4567\n\nInvalid latex syntax: $E = mc^$ and $$E = mc^$$",
"content": "This is the formula:\n\n$\\frac{e^{x_i}}{\\sum_{j=1}^{n}e^{x_j}}$\n\nGiven an input vector \\(\\mathbf{x} = [x_1, x_2, \\ldots, x_n]\\)\n\n\\[\ny_i = \\frac{e^{x_i}}{\\sum_{j=1}^n e^{x_j}}\n\\]\n\n$2x + y = z$\n\nCode block latex:\n```latex\n\\frac{e^{x_i}}{\\sum_{j=1}^{n}e^{x_j}}\n```\n\nTest dollar sign: $1234 $4567\n\nInvalid latex syntax: $E = mc^$ and $$E = mc^$$",
"timings": {
"prompt_n": 1,
"prompt_ms": 28.923,

36
examples/server/webui/src/App.tsx Normal file
View file

@ -0,0 +1,36 @@
import { HashRouter, Outlet, Route, Routes } from 'react-router';
import Header from './components/Header';
import Sidebar from './components/Sidebar';
import { AppContextProvider } from './utils/app.context';
import ChatScreen from './components/ChatScreen';
function App() {
return (
<HashRouter>
<div className="flex flex-row drawer lg:drawer-open">
<AppContextProvider>
<Routes>
<Route element={<AppLayout />}>
<Route path="/chat/:convId" element={<ChatScreen />} />
<Route path="*" element={<ChatScreen />} />
</Route>
</Routes>
</AppContextProvider>
</div>
</HashRouter>
);
}
function AppLayout() {
return (
<>
<Sidebar />
<div className="chat-screen drawer-content grow flex flex-col h-screen w-screen mx-auto px-4">
<Header />
<Outlet />
</div>
</>
);
}
export default App;

89
examples/server/webui/src/Config.ts Normal file
View file

@ -0,0 +1,89 @@
import daisyuiThemes from 'daisyui/src/theming/themes';
import { isNumeric } from './utils/misc';
export const isDev = import.meta.env.MODE === 'development';
// constants
export const BASE_URL = new URL('.', document.baseURI).href
.toString()
.replace(/\/$/, '');
export const CONFIG_DEFAULT = {
// Note: in order not to introduce breaking changes, please keep the same data type (number, string, etc) if you want to change the default value. Do not use null or undefined for default value.
apiKey: '',
systemMessage: 'You are a helpful assistant.',
showTokensPerSecond: false,
showThoughtInProgress: false,
excludeThoughtOnReq: true,
// make sure these default values are in sync with `common.h`
samplers: 'edkypmxt',
temperature: 0.8,
dynatemp_range: 0.0,
dynatemp_exponent: 1.0,
top_k: 40,
top_p: 0.95,
min_p: 0.05,
xtc_probability: 0.0,
xtc_threshold: 0.1,
typical_p: 1.0,
repeat_last_n: 64,
repeat_penalty: 1.0,
presence_penalty: 0.0,
frequency_penalty: 0.0,
dry_multiplier: 0.0,
dry_base: 1.75,
dry_allowed_length: 2,
dry_penalty_last_n: -1,
max_tokens: -1,
custom: '', // custom json-stringified object
};
export const CONFIG_INFO: Record<string, string> = {
apiKey: 'Set the API Key if you are using --api-key option for the server.',
systemMessage: 'The starting message that defines how model should behave.',
samplers:
'The order at which samplers are applied, in simplified way. Default is "dkypmxt": dry->top_k->typ_p->top_p->min_p->xtc->temperature',
temperature:
'Controls the randomness of the generated text by affecting the probability distribution of the output tokens. Higher = more random, lower = more focused.',
dynatemp_range:
'Addon for the temperature sampler. The added value to the range of dynamic temperature, which adjusts probabilities by entropy of tokens.',
dynatemp_exponent:
'Addon for the temperature sampler. Smoothes out the probability redistribution based on the most probable token.',
top_k: 'Keeps only k top tokens.',
top_p:
'Limits tokens to those that together have a cumulative probability of at least p',
min_p:
'Limits tokens based on the minimum probability for a token to be considered, relative to the probability of the most likely token.',
xtc_probability:
'XTC sampler cuts out top tokens; this parameter controls the chance of cutting tokens at all. 0 disables XTC.',
xtc_threshold:
'XTC sampler cuts out top tokens; this parameter controls the token probability that is required to cut that token.',
typical_p:
'Sorts and limits tokens based on the difference between log-probability and entropy.',
repeat_last_n: 'Last n tokens to consider for penalizing repetition',
repeat_penalty:
'Controls the repetition of token sequences in the generated text',
presence_penalty:
'Limits tokens based on whether they appear in the output or not.',
frequency_penalty:
'Limits tokens based on how often they appear in the output.',
dry_multiplier:
'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the DRY sampling multiplier.',
dry_base:
'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the DRY sampling base value.',
dry_allowed_length:
'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the allowed length for DRY sampling.',
dry_penalty_last_n:
'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets DRY penalty for the last n tokens.',
max_tokens: 'The maximum number of token per output.',
custom: '', // custom json-stringified object
};
// config keys having numeric value (i.e. temperature, top_k, top_p, etc)
export const CONFIG_NUMERIC_KEYS = Object.entries(CONFIG_DEFAULT)
.filter((e) => isNumeric(e[1]))
.map((e) => e[0]);
// list of themes supported by daisyui
export const THEMES = ['light', 'dark']
// make sure light & dark are always at the beginning
.concat(
Object.keys(daisyuiThemes).filter((t) => t !== 'light' && t !== 'dark')
);

229
examples/server/webui/src/components/ChatMessage.tsx Normal file
View file

@ -0,0 +1,229 @@
import { useMemo, useState } from 'react';
import { useAppContext } from '../utils/app.context';
import { Message, PendingMessage } from '../utils/types';
import { classNames } from '../utils/misc';
import MarkdownDisplay, { CopyButton } from './MarkdownDisplay';
interface SplitMessage {
content: PendingMessage['content'];
thought?: string;
isThinking?: boolean;
}
export default function ChatMessage({
msg,
id,
scrollToBottom,
isPending,
}: {
msg: Message | PendingMessage;
id?: string;
scrollToBottom: (requiresNearBottom: boolean) => void;
isPending?: boolean;
}) {
const { viewingConversation, replaceMessageAndGenerate, config } =
useAppContext();
const [editingContent, setEditingContent] = useState<string | null>(null);
const timings = useMemo(
() =>
msg.timings
? {
...msg.timings,
prompt_per_second:
(msg.timings.prompt_n / msg.timings.prompt_ms) * 1000,
predicted_per_second:
(msg.timings.predicted_n / msg.timings.predicted_ms) * 1000,
}
: null,
[msg.timings]
);
// for reasoning model, we split the message into content and thought
// TODO: implement this as remark/rehype plugin in the future
const { content, thought, isThinking }: SplitMessage = useMemo(() => {
if (msg.content === null || msg.role !== 'assistant') {
return { content: msg.content };
}
let actualContent = '';
let thought = '';
let isThinking = false;
let thinkSplit = msg.content.split('<think>', 2);
actualContent += thinkSplit[0];
while (thinkSplit[1] !== undefined) {
// <think> tag found
thinkSplit = thinkSplit[1].split('</think>', 2);
thought += thinkSplit[0];
isThinking = true;
if (thinkSplit[1] !== undefined) {
// </think> closing tag found
isThinking = false;
thinkSplit = thinkSplit[1].split('<think>', 2);
actualContent += thinkSplit[0];
}
}
return { content: actualContent, thought, isThinking };
}, [msg]);
if (!viewingConversation) return null;
const regenerate = async () => {
replaceMessageAndGenerate(viewingConversation.id, msg.id, undefined, () =>
scrollToBottom(true)
);
};
return (
<div className="group" id={id}>
<div
className={classNames({
chat: true,
'chat-start': msg.role !== 'user',
'chat-end': msg.role === 'user',
})}
>
<div
className={classNames({
'chat-bubble markdown': true,
'chat-bubble-base-300': msg.role !== 'user',
})}
>
{/* textarea for editing message */}
{editingContent !== null && (
<>
<textarea
dir="auto"
className="textarea textarea-bordered bg-base-100 text-base-content w-[calc(90vw-8em)] lg:w-96"
value={editingContent}
onChange={(e) => setEditingContent(e.target.value)}
></textarea>
<br />
<button
className="btn btn-ghost mt-2 mr-2"
onClick={() => setEditingContent(null)}
>
Cancel
</button>
<button
className="btn mt-2"
onClick={() =>
replaceMessageAndGenerate(
viewingConversation.id,
msg.id,
editingContent
)
}
>
Submit
</button>
</>
)}
{/* not editing content, render message */}
{editingContent === null && (
<>
{content === null ? (
<>
{/* show loading dots for pending message */}
<span className="loading loading-dots loading-md"></span>
</>
) : (
<>
{/* render message as markdown */}
<div dir="auto">
{thought && (
<details
className="collapse bg-base-200 collapse-arrow mb-4"
open={isThinking && config.showThoughtInProgress}
>
<summary className="collapse-title">
{isPending && isThinking ? (
<span>
<span
v-if="isGenerating"
className="loading loading-spinner loading-md mr-2"
style={{ verticalAlign: 'middle' }}
></span>
<b>Thinking</b>
</span>
) : (
<b>Thought Process</b>
)}
</summary>
<div className="collapse-content">
<MarkdownDisplay content={thought} />
</div>
</details>
)}
<MarkdownDisplay content={content} />
</div>
</>
)}
{/* render timings if enabled */}
{timings && config.showTokensPerSecond && (
<div className="dropdown dropdown-hover dropdown-top mt-2">
<div
tabIndex={0}
role="button"
className="cursor-pointer font-semibold text-sm opacity-60"
>
Speed: {timings.predicted_per_second.toFixed(1)} t/s
</div>
<div className="dropdown-content bg-base-100 z-10 w-64 p-2 shadow mt-4">
<b>Prompt</b>
<br />- Tokens: {timings.prompt_n}
<br />- Time: {timings.prompt_ms} ms
<br />- Speed: {timings.prompt_per_second.toFixed(1)} t/s
<br />
<b>Generation</b>
<br />- Tokens: {timings.predicted_n}
<br />- Time: {timings.predicted_ms} ms
<br />- Speed: {timings.predicted_per_second.toFixed(1)} t/s
<br />
</div>
</div>
)}
</>
)}
</div>
</div>
{/* actions for each message */}
{msg.content !== null && (
<div
className={classNames({
'mx-4 mt-2 mb-2': true,
'text-right': msg.role === 'user',
})}
>
{/* user message */}
{msg.role === 'user' && (
<button
className="badge btn-mini show-on-hover"
onClick={() => setEditingContent(msg.content)}
disabled={msg.content === null}
>
✍️ Edit
</button>
)}
{/* assistant message */}
{msg.role === 'assistant' && (
<>
{!isPending && (
<button
className="badge btn-mini show-on-hover mr-2"
onClick={regenerate}
disabled={msg.content === null}
>
🔄 Regenerate
</button>
)}
<CopyButton
className="badge btn-mini show-on-hover mr-2"
content={msg.content}
/>
</>
)}
</div>
)}
</div>
);
}

123
examples/server/webui/src/components/ChatScreen.tsx Normal file
View file

@ -0,0 +1,123 @@
import { useEffect, useRef, useState } from 'react';
import { useAppContext } from '../utils/app.context';
import StorageUtils from '../utils/storage';
import { useNavigate } from 'react-router';
import ChatMessage from './ChatMessage';
import { PendingMessage } from '../utils/types';
export default function ChatScreen() {
const {
viewingConversation,
sendMessage,
isGenerating,
stopGenerating,
pendingMessages,
} = useAppContext();
const [inputMsg, setInputMsg] = useState('');
const containerRef = useRef<HTMLDivElement>(null);
const navigate = useNavigate();
const currConvId = viewingConversation?.id ?? '';
const pendingMsg: PendingMessage | undefined = pendingMessages[currConvId];
const scrollToBottom = (requiresNearBottom: boolean) => {
if (!containerRef.current) return;
const msgListElem = containerRef.current;
const spaceToBottom =
msgListElem.scrollHeight -
msgListElem.scrollTop -
msgListElem.clientHeight;
if (!requiresNearBottom || spaceToBottom < 50) {
setTimeout(
() => msgListElem.scrollTo({ top: msgListElem.scrollHeight }),
1
);
}
};
// scroll to bottom when conversation changes
useEffect(() => {
scrollToBottom(false);
}, [viewingConversation?.id]);
const sendNewMessage = async () => {
if (inputMsg.trim().length === 0 || isGenerating(currConvId)) return;
const convId = viewingConversation?.id ?? StorageUtils.getNewConvId();
const lastInpMsg = inputMsg;
setInputMsg('');
if (!viewingConversation) {
// if user is creating a new conversation, redirect to the new conversation
navigate(`/chat/${convId}`);
}
scrollToBottom(false);
// auto scroll as message is being generated
const onChunk = () => scrollToBottom(true);
if (!(await sendMessage(convId, inputMsg, onChunk))) {
// restore the input message if failed
setInputMsg(lastInpMsg);
}
};
return (
<>
{/* chat messages */}
<div
id="messages-list"
className="flex flex-col grow overflow-y-auto"
ref={containerRef}
>
<div className="mt-auto flex justify-center">
{/* placeholder to shift the message to the bottom */}
{viewingConversation ? '' : 'Send a message to start'}
</div>
{viewingConversation?.messages.map((msg) => (
<ChatMessage key={msg.id} msg={msg} scrollToBottom={scrollToBottom} />
))}
{pendingMsg && (
<ChatMessage
msg={pendingMsg}
scrollToBottom={scrollToBottom}
isPending
id="pending-msg"
/>
)}
</div>
{/* chat input */}
<div className="flex flex-row items-center mt-8 mb-6">
<textarea
className="textarea textarea-bordered w-full"
placeholder="Type a message (Shift+Enter to add a new line)"
value={inputMsg}
onChange={(e) => setInputMsg(e.target.value)}
onKeyDown={(e) => {
if (e.key === 'Enter' && e.shiftKey) return;
if (e.key === 'Enter' && !e.shiftKey) {
e.preventDefault();
sendNewMessage();
}
}}
id="msg-input"
dir="auto"
></textarea>
{isGenerating(currConvId) ? (
<button
className="btn btn-neutral ml-2"
onClick={() => stopGenerating(currConvId)}
>
Stop
</button>
) : (
<button
className="btn btn-primary ml-2"
onClick={sendNewMessage}
disabled={inputMsg.trim().length === 0}
>
Send
</button>
)}
</div>
</>
);
}

182
examples/server/webui/src/components/Header.tsx Normal file
View file

@ -0,0 +1,182 @@
import { useEffect, useState } from 'react';
import StorageUtils from '../utils/storage';
import { useAppContext } from '../utils/app.context';
import { classNames } from '../utils/misc';
import daisyuiThemes from 'daisyui/src/theming/themes';
import { THEMES } from '../Config';
import { useNavigate } from 'react-router';
import SettingDialog from './SettingDialog';
export default function Header() {
const navigate = useNavigate();
const [selectedTheme, setSelectedTheme] = useState(StorageUtils.getTheme());
const [showSettingDialog, setShowSettingDialog] = useState(false);
const setTheme = (theme: string) => {
StorageUtils.setTheme(theme);
setSelectedTheme(theme);
};
useEffect(() => {
document.body.setAttribute('data-theme', selectedTheme);
document.body.setAttribute(
'data-color-scheme',
// @ts-expect-error daisyuiThemes complains about index type, but it should work
daisyuiThemes[selectedTheme]?.['color-scheme'] ?? 'auto'
);
}, [selectedTheme]);
const { isGenerating, viewingConversation } = useAppContext();
const isCurrConvGenerating = isGenerating(viewingConversation?.id ?? '');
const removeConversation = () => {
if (isCurrConvGenerating || !viewingConversation) return;
const convId = viewingConversation.id;
if (window.confirm('Are you sure to delete this conversation?')) {
StorageUtils.remove(convId);
navigate('/');
}
};
const downloadConversation = () => {
if (isCurrConvGenerating || !viewingConversation) return;
const convId = viewingConversation.id;
const conversationJson = JSON.stringify(viewingConversation, null, 2);
const blob = new Blob([conversationJson], { type: 'application/json' });
const url = URL.createObjectURL(blob);
const a = document.createElement('a');
a.href = url;
a.download = `conversation_${convId}.json`;
document.body.appendChild(a);
a.click();
document.body.removeChild(a);
URL.revokeObjectURL(url);
};
return (
<div className="flex flex-row items-center mt-6 mb-6">
{/* open sidebar button */}
<label htmlFor="toggle-drawer" className="btn btn-ghost lg:hidden">
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
fill="currentColor"
className="bi bi-list"
viewBox="0 0 16 16"
>
<path
fillRule="evenodd"
d="M2.5 12a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5m0-4a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5m0-4a.5.5 0 0 1 .5-.5h10a.5.5 0 0 1 0 1H3a.5.5 0 0 1-.5-.5"
/>
</svg>
</label>
<div className="grow text-2xl font-bold ml-2">llama.cpp</div>
{/* action buttons (top right) */}
<div className="flex items-center">
<div v-if="messages.length > 0" className="dropdown dropdown-end">
{/* "..." button */}
<button
tabIndex={0}
role="button"
className="btn m-1"
disabled={isCurrConvGenerating}
>
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
fill="currentColor"
className="bi bi-three-dots-vertical"
viewBox="0 0 16 16"
>
<path d="M9.5 13a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0m0-5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0m0-5a1.5 1.5 0 1 1-3 0 1.5 1.5 0 0 1 3 0" />
</svg>
</button>
{/* dropdown menu */}
<ul
tabIndex={0}
className="dropdown-content menu bg-base-100 rounded-box z-[1] w-52 p-2 shadow"
>
<li onClick={downloadConversation}>
<a>Download</a>
</li>
<li className="text-error" onClick={removeConversation}>
<a>Delete</a>
</li>
</ul>
</div>
<div className="tooltip tooltip-bottom" data-tip="Settings">
<button className="btn" onClick={() => setShowSettingDialog(true)}>
{/* settings button */}
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
fill="currentColor"
className="bi bi-gear"
viewBox="0 0 16 16"
>
<path d="M8 4.754a3.246 3.246 0 1 0 0 6.492 3.246 3.246 0 0 0 0-6.492M5.754 8a2.246 2.246 0 1 1 4.492 0 2.246 2.246 0 0 1-4.492 0" />
<path d="M9.796 1.343c-.527-1.79-3.065-1.79-3.592 0l-.094.319a.873.873 0 0 1-1.255.52l-.292-.16c-1.64-.892-3.433.902-2.54 2.541l.159.292a.873.873 0 0 1-.52 1.255l-.319.094c-1.79.527-1.79 3.065 0 3.592l.319.094a.873.873 0 0 1 .52 1.255l-.16.292c-.892 1.64.901 3.434 2.541 2.54l.292-.159a.873.873 0 0 1 1.255.52l.094.319c.527 1.79 3.065 1.79 3.592 0l.094-.319a.873.873 0 0 1 1.255-.52l.292.16c1.64.893 3.434-.902 2.54-2.541l-.159-.292a.873.873 0 0 1 .52-1.255l.319-.094c1.79-.527 1.79-3.065 0-3.592l-.319-.094a.873.873 0 0 1-.52-1.255l.16-.292c.893-1.64-.902-3.433-2.541-2.54l-.292.159a.873.873 0 0 1-1.255-.52zm-2.633.283c.246-.835 1.428-.835 1.674 0l.094.319a1.873 1.873 0 0 0 2.693 1.115l.291-.16c.764-.415 1.6.42 1.184 1.185l-.159.292a1.873 1.873 0 0 0 1.116 2.692l.318.094c.835.246.835 1.428 0 1.674l-.319.094a1.873 1.873 0 0 0-1.115 2.693l.16.291c.415.764-.42 1.6-1.185 1.184l-.291-.159a1.873 1.873 0 0 0-2.693 1.116l-.094.318c-.246.835-1.428.835-1.674 0l-.094-.319a1.873 1.873 0 0 0-2.692-1.115l-.292.16c-.764.415-1.6-.42-1.184-1.185l.159-.291A1.873 1.873 0 0 0 1.945 8.93l-.319-.094c-.835-.246-.835-1.428 0-1.674l.319-.094A1.873 1.873 0 0 0 3.06 4.377l-.16-.292c-.415-.764.42-1.6 1.185-1.184l.292.159a1.873 1.873 0 0 0 2.692-1.115z" />
</svg>
</button>
</div>
{/* theme controller is copied from https://daisyui.com/components/theme-controller/ */}
<div className="tooltip tooltip-bottom" data-tip="Themes">
<div className="dropdown dropdown-end dropdown-bottom">
<div tabIndex={0} role="button" className="btn m-1">
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
fill="currentColor"
className="bi bi-palette2"
viewBox="0 0 16 16"
>
<path d="M0 .5A.5.5 0 0 1 .5 0h5a.5.5 0 0 1 .5.5v5.277l4.147-4.131a.5.5 0 0 1 .707 0l3.535 3.536a.5.5 0 0 1 0 .708L10.261 10H15.5a.5.5 0 0 1 .5.5v5a.5.5 0 0 1-.5.5H3a3 3 0 0 1-2.121-.879A3 3 0 0 1 0 13.044m6-.21 7.328-7.3-2.829-2.828L6 7.188zM4.5 13a1.5 1.5 0 1 0-3 0 1.5 1.5 0 0 0 3 0M15 15v-4H9.258l-4.015 4zM0 .5v12.495zm0 12.495V13z" />
</svg>
</div>
<ul
tabIndex={0}
className="dropdown-content bg-base-300 rounded-box z-[1] w-52 p-2 shadow-2xl h-80 overflow-y-auto"
>
<li>
<button
className={classNames({
'btn btn-sm btn-block btn-ghost justify-start': true,
'btn-active': selectedTheme === 'auto',
})}
onClick={() => setTheme('auto')}
>
auto
</button>
</li>
{THEMES.map((theme) => (
<li key={theme}>
<input
type="radio"
name="theme-dropdown"
className="theme-controller btn btn-sm btn-block btn-ghost justify-start"
aria-label={theme}
value={theme}
checked={selectedTheme === theme}
onChange={(e) => e.target.checked && setTheme(theme)}
/>
</li>
))}
</ul>
</div>
</div>
</div>
<SettingDialog
show={showSettingDialog}
onClose={() => setShowSettingDialog(false)}
/>
</div>
);
}

254
examples/server/webui/src/components/MarkdownDisplay.tsx Normal file
View file

@ -0,0 +1,254 @@
import React, { useMemo, useState } from 'react';
import Markdown, { ExtraProps } from 'react-markdown';
import remarkGfm from 'remark-gfm';
import rehypeHightlight from 'rehype-highlight';
import rehypeKatex from 'rehype-katex';
import remarkMath from 'remark-math';
import remarkBreaks from 'remark-breaks';
import 'katex/dist/katex.min.css';
import { classNames, copyStr } from '../utils/misc';
import { ElementContent, Root } from 'hast';
import { visit } from 'unist-util-visit';
export default function MarkdownDisplay({ content }: { content: string }) {
const preprocessedContent = useMemo(
() => preprocessLaTeX(content),
[content]
);
return (
<Markdown
remarkPlugins={[remarkGfm, remarkMath, remarkBreaks]}
rehypePlugins={[rehypeHightlight, rehypeKatex, rehypeCustomCopyButton]}
components={{
button: (props) => (
<CopyCodeButton {...props} origContent={preprocessedContent} />
),
}}
>
{preprocessedContent}
</Markdown>
);
}
const CopyCodeButton: React.ElementType<
React.ClassAttributes<HTMLButtonElement> &
React.HTMLAttributes<HTMLButtonElement> &
ExtraProps & { origContent: string }
> = ({ node, origContent }) => {
const startOffset = node?.position?.start.offset ?? 0;
const endOffset = node?.position?.end.offset ?? 0;
const copiedContent = useMemo(
() =>
origContent
.substring(startOffset, endOffset)
.replace(/^```[^\n]+\n/g, '')
.replace(/```$/g, ''),
[origContent, startOffset, endOffset]
);
return (
<div
className={classNames({
'text-right sticky top-4 mb-2 mr-2 h-0': true,
'display-none': !node?.position,
})}
>
<CopyButton className="badge btn-mini" content={copiedContent} />
</div>
);
};
export const CopyButton = ({
content,
className,
}: {
content: string;
className?: string;
}) => {
const [copied, setCopied] = useState(false);
return (
<button
className={className}
onClick={() => {
copyStr(content);
setCopied(true);
}}
onMouseLeave={() => setCopied(false)}
>
{copied ? 'Copied!' : '📋 Copy'}
</button>
);
};
/**
* This injects the "button" element before each "pre" element.
* The actual button will be replaced with a react component in the MarkdownDisplay.
* We don't replace "pre" node directly because it will cause the node to re-render, which causes this bug: https://github.com/ggerganov/llama.cpp/issues/9608
*/
function rehypeCustomCopyButton() {
return function (tree: Root) {
visit(tree, 'element', function (node) {
if (node.tagName === 'pre' && !node.properties.visited) {
const preNode = { ...node };
// replace current node
preNode.properties.visited = 'true';
node.tagName = 'div';
node.properties = {
className: 'relative my-4',
};
// add node for button
const btnNode: ElementContent = {
type: 'element',
tagName: 'button',
properties: {},
children: [],
position: node.position,
};
node.children = [btnNode, preNode];
}
});
};
}
/**
* The part below is copied and adapted from:
* https://github.com/danny-avila/LibreChat/blob/main/client/src/utils/latex.ts
* (MIT License)
*/
// Regex to check if the processed content contains any potential LaTeX patterns
const containsLatexRegex =
/\\\(.*?\\\)|\\\[.*?\\\]|\$.*?\$|\\begin\{equation\}.*?\\end\{equation\}/;
// Regex for inline and block LaTeX expressions
const inlineLatex = new RegExp(/\\\((.+?)\\\)/, 'g');
const blockLatex = new RegExp(/\\\[(.*?[^\\])\\\]/, 'gs');
// Function to restore code blocks
const restoreCodeBlocks = (content: string, codeBlocks: string[]) => {
return content.replace(
/<<CODE_BLOCK_(\d+)>>/g,
(_, index) => codeBlocks[index]
);
};
// Regex to identify code blocks and inline code
const codeBlockRegex = /(```[\s\S]*?```|`.*?`)/g;
export const processLaTeX = (_content: string) => {
let content = _content;
// Temporarily replace code blocks and inline code with placeholders
const codeBlocks: string[] = [];
let index = 0;
content = content.replace(codeBlockRegex, (match) => {
codeBlocks[index] = match;
return `<<CODE_BLOCK_${index++}>>`;
});
// Escape dollar signs followed by a digit or space and digit
let processedContent = content.replace(/(\$)(?=\s?\d)/g, '\\$');
// If no LaTeX patterns are found, restore code blocks and return the processed content
if (!containsLatexRegex.test(processedContent)) {
return restoreCodeBlocks(processedContent, codeBlocks);
}
// Convert LaTeX expressions to a markdown compatible format
processedContent = processedContent
.replace(inlineLatex, (_: string, equation: string) => `$${equation}$`) // Convert inline LaTeX
.replace(blockLatex, (_: string, equation: string) => `$$${equation}$$`); // Convert block LaTeX
// Restore code blocks
return restoreCodeBlocks(processedContent, codeBlocks);
};
/**
* Preprocesses LaTeX content by replacing delimiters and escaping certain characters.
*
* @param content The input string containing LaTeX expressions.
* @returns The processed string with replaced delimiters and escaped characters.
*/
export function preprocessLaTeX(content: string): string {
// Step 1: Protect code blocks
const codeBlocks: string[] = [];
content = content.replace(/(```[\s\S]*?```|`[^`\n]+`)/g, (_, code) => {
codeBlocks.push(code);
return `<<CODE_BLOCK_${codeBlocks.length - 1}>>`;
});
// Step 2: Protect existing LaTeX expressions
const latexExpressions: string[] = [];
// Protect block math ($$...$$), \[...\], and \(...\) as before.
content = content.replace(
/(\$\$[\s\S]*?\$\$|\\\[[\s\S]*?\\\]|\\\(.*?\\\))/g,
(match) => {
latexExpressions.push(match);
return `<<LATEX_${latexExpressions.length - 1}>>`;
}
);
// Protect inline math ($...$) only if it does NOT match a currency pattern.
// We assume a currency pattern is one where the inner content is purely numeric (with optional decimals).
content = content.replace(/\$([^$]+)\$/g, (match, inner) => {
if (/^\s*\d+(?:\.\d+)?\s*$/.test(inner)) {
// This looks like a currency value (e.g. "$123" or "$12.34"),
// so don't protect it.
return match;
} else {
// Otherwise, treat it as a LaTeX expression.
latexExpressions.push(match);
return `<<LATEX_${latexExpressions.length - 1}>>`;
}
});
// Step 3: Escape dollar signs that are likely currency indicators.
// (Now that inline math is protected, this will only escape dollars not already protected)
content = content.replace(/\$(?=\d)/g, '\\$');
// Step 4: Restore LaTeX expressions
content = content.replace(
/<<LATEX_(\d+)>>/g,
(_, index) => latexExpressions[parseInt(index)]
);
// Step 5: Restore code blocks
content = content.replace(
/<<CODE_BLOCK_(\d+)>>/g,
(_, index) => codeBlocks[parseInt(index)]
);
// Step 6: Apply additional escaping functions
content = escapeBrackets(content);
content = escapeMhchem(content);
return content;
}
export function escapeBrackets(text: string): string {
const pattern =
/(```[\S\s]*?```|`.*?`)|\\\[([\S\s]*?[^\\])\\]|\\\((.*?)\\\)/g;
return text.replace(
pattern,
(
match: string,
codeBlock: string | undefined,
squareBracket: string | undefined,
roundBracket: string | undefined
): string => {
if (codeBlock != null) {
return codeBlock;
} else if (squareBracket != null) {
return `$$${squareBracket}$$`;
} else if (roundBracket != null) {
return `$${roundBracket}$`;
}
return match;
}
);
}
export function escapeMhchem(text: string) {
return text.replaceAll('$\\ce{', '$\\\\ce{').replaceAll('$\\pu{', '$\\\\pu{');
}

306
examples/server/webui/src/components/SettingDialog.tsx Normal file
View file

@ -0,0 +1,306 @@
import { useState } from 'react';
import { useAppContext } from '../utils/app.context';
import { CONFIG_DEFAULT, CONFIG_INFO } from '../Config';
import { isDev } from '../Config';
import StorageUtils from '../utils/storage';
type SettKey = keyof typeof CONFIG_DEFAULT;
const COMMON_SAMPLER_KEYS: SettKey[] = [
'temperature',
'top_k',
'top_p',
'min_p',
'max_tokens',
];
const OTHER_SAMPLER_KEYS: SettKey[] = [
'dynatemp_range',
'dynatemp_exponent',
'typical_p',
'xtc_probability',
'xtc_threshold',
];
const PENALTY_KEYS: SettKey[] = [
'repeat_last_n',
'repeat_penalty',
'presence_penalty',
'frequency_penalty',
'dry_multiplier',
'dry_base',
'dry_allowed_length',
'dry_penalty_last_n',
];
export default function SettingDialog({
show,
onClose,
}: {
show: boolean;
onClose: () => void;
}) {
const { config, saveConfig } = useAppContext();
// clone the config object to prevent direct mutation
const [localConfig, setLocalConfig] = useState<typeof CONFIG_DEFAULT>(
JSON.parse(JSON.stringify(config))
);
const resetConfig = () => {
if (window.confirm('Are you sure to reset all settings?')) {
setLocalConfig(CONFIG_DEFAULT);
}
};
const handleSave = () => {
saveConfig(localConfig);
onClose();
};
const debugImportDemoConv = async () => {
const res = await fetch('/demo-conversation.json');
const demoConv = await res.json();
StorageUtils.remove(demoConv.id);
for (const msg of demoConv.messages) {
StorageUtils.appendMsg(demoConv.id, msg);
}
onClose();
};
return (
<dialog className={`modal ${show ? 'modal-open' : ''}`}>
<div className="modal-box">
<h3 className="text-lg font-bold mb-6">Settings</h3>
<div className="h-[calc(90vh-12rem)] overflow-y-auto">
<p className="opacity-40 mb-6">
Settings below are saved in browser's localStorage
</p>
<SettingsModalShortInput
configKey="apiKey"
configDefault={CONFIG_DEFAULT}
value={localConfig.apiKey}
onChange={(value) =>
setLocalConfig({ ...localConfig, apiKey: value })
}
/>
<label className="form-control mb-2">
<div className="label">
System Message (will be disabled if left empty)
</div>
<textarea
className="textarea textarea-bordered h-24"
placeholder={`Default: ${CONFIG_DEFAULT.systemMessage}`}
value={localConfig.systemMessage}
onChange={(e) =>
setLocalConfig({
...localConfig,
systemMessage: e.target.value,
})
}
/>
</label>
{COMMON_SAMPLER_KEYS.map((key) => (
<SettingsModalShortInput
key={key}
configKey={key}
configDefault={CONFIG_DEFAULT}
value={localConfig[key]}
onChange={(value) =>
setLocalConfig({ ...localConfig, [key]: value })
}
/>
))}
<details className="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary className="collapse-title font-bold">
Other sampler settings
</summary>
<div className="collapse-content">
<SettingsModalShortInput
label="Samplers queue"
configKey="samplers"
configDefault={CONFIG_DEFAULT}
value={localConfig.samplers}
onChange={(value) =>
setLocalConfig({ ...localConfig, samplers: value })
}
/>
{OTHER_SAMPLER_KEYS.map((key) => (
<SettingsModalShortInput
key={key}
configKey={key}
configDefault={CONFIG_DEFAULT}
value={localConfig[key]}
onChange={(value) =>
setLocalConfig({ ...localConfig, [key]: value })
}
/>
))}
</div>
</details>
<details className="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary className="collapse-title font-bold">
Penalties settings
</summary>
<div className="collapse-content">
{PENALTY_KEYS.map((key) => (
<SettingsModalShortInput
key={key}
configKey={key}
configDefault={CONFIG_DEFAULT}
value={localConfig[key]}
onChange={(value) =>
setLocalConfig({ ...localConfig, [key]: value })
}
/>
))}
</div>
</details>
<details className="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary className="collapse-title font-bold">
Reasoning models
</summary>
<div className="collapse-content">
<div className="flex flex-row items-center mb-2">
<input
type="checkbox"
className="checkbox"
checked={localConfig.showThoughtInProgress}
onChange={(e) =>
setLocalConfig({
...localConfig,
showThoughtInProgress: e.target.checked,
})
}
/>
<span className="ml-4">
Expand though process by default for generating message
</span>
</div>
<div className="flex flex-row items-center mb-2">
<input
type="checkbox"
className="checkbox"
checked={localConfig.excludeThoughtOnReq}
onChange={(e) =>
setLocalConfig({
...localConfig,
excludeThoughtOnReq: e.target.checked,
})
}
/>
<span className="ml-4">
Exclude thought process when sending request to API
(Recommended for DeepSeek-R1)
</span>
</div>
</div>
</details>
<details className="collapse collapse-arrow bg-base-200 mb-2 overflow-visible">
<summary className="collapse-title font-bold">
Advanced config
</summary>
<div className="collapse-content">
{/* this button only shows in dev mode, used to import a demo conversation to test message rendering */}
{isDev && (
<div className="flex flex-row items-center mb-2">
<button className="btn" onClick={debugImportDemoConv}>
(debug) Import demo conversation
</button>
</div>
)}
<div className="flex flex-row items-center mb-2">
<input
type="checkbox"
className="checkbox"
checked={localConfig.showTokensPerSecond}
onChange={(e) =>
setLocalConfig({
...localConfig,
showTokensPerSecond: e.target.checked,
})
}
/>
<span className="ml-4">Show tokens per second</span>
</div>
<label className="form-control mb-2">
<div className="label inline">
Custom JSON config (For more info, refer to{' '}
<a
className="underline"
href="https://github.com/ggerganov/llama.cpp/blob/master/examples/server/README.md"
target="_blank"
rel="noopener noreferrer"
>
server documentation
</a>
)
</div>
<textarea
className="textarea textarea-bordered h-24"
placeholder='Example: { "mirostat": 1, "min_p": 0.1 }'
value={localConfig.custom}
onChange={(e) =>
setLocalConfig({ ...localConfig, custom: e.target.value })
}
/>
</label>
</div>
</details>
</div>
<div className="modal-action">
<button className="btn" onClick={resetConfig}>
Reset to default
</button>
<button className="btn" onClick={onClose}>
Close
</button>
<button className="btn btn-primary" onClick={handleSave}>
Save
</button>
</div>
</div>
</dialog>
);
}
function SettingsModalShortInput({
configKey,
configDefault,
value,
onChange,
label,
}: {
configKey: SettKey;
configDefault: typeof CONFIG_DEFAULT;
// eslint-disable-next-line @typescript-eslint/no-explicit-any
value: any;
onChange: (value: string) => void;
label?: string;
}) {
return (
<label className="input input-bordered join-item grow flex items-center gap-2 mb-2">
<div className="dropdown dropdown-hover">
<div tabIndex={0} role="button" className="font-bold">
{label || configKey}
</div>
<div className="dropdown-content menu bg-base-100 rounded-box z-10 w-64 p-2 shadow mt-4">
{CONFIG_INFO[configKey] ?? '(no help message available)'}
</div>
</div>
<input
type="text"
className="grow"
placeholder={`Default: ${configDefault[configKey] || 'none'}`}
value={value}
onChange={(e) => onChange(e.target.value)}
/>
</label>
);
}

95
examples/server/webui/src/components/Sidebar.tsx Normal file
View file

@ -0,0 +1,95 @@
import { useEffect, useMemo, useState } from 'react';
import { classNames } from '../utils/misc';
import { Conversation } from '../utils/types';
import StorageUtils from '../utils/storage';
import { useNavigate, useParams } from 'react-router';
export default function Sidebar() {
const params = useParams();
const navigate = useNavigate();
const currConv = useMemo(
() => StorageUtils.getOneConversation(params.convId ?? ''),
[params.convId]
);
const [conversations, setConversations] = useState<Conversation[]>([]);
useEffect(() => {
const handleConversationChange = () => {
setConversations(StorageUtils.getAllConversations());
};
StorageUtils.onConversationChanged(handleConversationChange);
handleConversationChange();
return () => {
StorageUtils.offConversationChanged(handleConversationChange);
};
}, []);
return (
<>
<input
id="toggle-drawer"
type="checkbox"
className="drawer-toggle"
defaultChecked
/>
<div className="drawer-side h-screen lg:h-screen z-50 lg:max-w-64">
<label
htmlFor="toggle-drawer"
aria-label="close sidebar"
className="drawer-overlay"
></label>
<div className="flex flex-col bg-base-200 min-h-full max-w-64 py-4 px-4">
<div className="flex flex-row items-center justify-between mb-4 mt-4">
<h2 className="font-bold ml-4">Conversations</h2>
{/* close sidebar button */}
<label htmlFor="toggle-drawer" className="btn btn-ghost lg:hidden">
<svg
xmlns="http://www.w3.org/2000/svg"
width="16"
height="16"
fill="currentColor"
className="bi bi-arrow-bar-left"
viewBox="0 0 16 16"
>
<path
fillRule="evenodd"
d="M12.5 15a.5.5 0 0 1-.5-.5v-13a.5.5 0 0 1 1 0v13a.5.5 0 0 1-.5.5M10 8a.5.5 0 0 1-.5.5H3.707l2.147 2.146a.5.5 0 0 1-.708.708l-3-3a.5.5 0 0 1 0-.708l3-3a.5.5 0 1 1 .708.708L3.707 7.5H9.5a.5.5 0 0 1 .5.5"
/>
</svg>
</label>
</div>
{/* list of conversations */}
<div
className={classNames({
'btn btn-ghost justify-start': true,
'btn-active': !currConv,
})}
onClick={() => navigate('/')}
>
+ New conversation
</div>
{conversations.map((conv) => (
<div
key={conv.id}
className={classNames({
'btn btn-ghost justify-start font-normal': true,
'btn-active': conv.id === currConv?.id,
})}
onClick={() => navigate(`/chat/${conv.id}`)}
dir="auto"
>
<span className="truncate">{conv.messages[0].content}</span>
</div>
))}
<div className="text-center text-xs opacity-40 mt-auto mx-4">
Conversations are saved to browser's localStorage
</div>
</div>
</div>
</>
);
}

60
examples/server/webui/src/highlight-config.js
View file

@ -1,60 +0,0 @@
import hljs from 'highlight.js/lib/core';
// only import commonly used languages to reduce bundle size
import python from 'highlight.js/lib/languages/python';
import javascript from 'highlight.js/lib/languages/javascript';
import json from 'highlight.js/lib/languages/json';
import bash from 'highlight.js/lib/languages/bash';
import yaml from 'highlight.js/lib/languages/yaml';
import markdown from 'highlight.js/lib/languages/markdown';
import scss from 'highlight.js/lib/languages/scss';
import xml from 'highlight.js/lib/languages/xml';
import ruby from 'highlight.js/lib/languages/ruby';
import go from 'highlight.js/lib/languages/go';
import java from 'highlight.js/lib/languages/java';
import rust from 'highlight.js/lib/languages/rust';
import scala from 'highlight.js/lib/languages/scala';
import cpp from 'highlight.js/lib/languages/cpp';
import csharp from 'highlight.js/lib/languages/csharp';
import swift from 'highlight.js/lib/languages/swift';
import dart from 'highlight.js/lib/languages/dart';
import elixir from 'highlight.js/lib/languages/elixir';
import kotlin from 'highlight.js/lib/languages/kotlin';
import lua from 'highlight.js/lib/languages/lua';
import php from 'highlight.js/lib/languages/php';
import latex from 'highlight.js/lib/languages/latex';
hljs.registerLanguage('python', python);
hljs.registerLanguage('javascript', javascript);
hljs.registerLanguage('json', json);
hljs.registerLanguage('yaml', yaml);
hljs.registerLanguage('markdown', markdown);
hljs.registerLanguage('xml', xml);
hljs.registerLanguage('ruby', ruby);
hljs.registerLanguage('go', go);
hljs.registerLanguage('java', java);
hljs.registerLanguage('rust', rust);
hljs.registerLanguage('scala', scala);
hljs.registerLanguage('csharp', csharp);
hljs.registerLanguage('swift', swift);
hljs.registerLanguage('dart', dart);
hljs.registerLanguage('elixir', elixir);
hljs.registerLanguage('kotlin', kotlin);
hljs.registerLanguage('lua', lua);
hljs.registerLanguage('php', php);
hljs.registerLanguage('latex', latex);
// reuse some languages to further reduce bundle size
hljs.registerLanguage('shell', bash);
hljs.registerLanguage('bash', bash);
hljs.registerLanguage('sh', bash);
hljs.registerLanguage('css', scss);
hljs.registerLanguage('scss', scss);
hljs.registerLanguage('c', cpp);
hljs.registerLanguage('cpp', cpp);
export default hljs;

25
examples/server/webui/src/styles.scss → examples/server/webui/src/index.scss
View file

@ -1,15 +1,28 @@
@use "sass:meta";
@use 'sass:meta';
@tailwind base;
@tailwind components;
@tailwind utilities;
.markdown {
h1, h2, h3, h4, h5, h6, ul, ol, li { all: revert; }
h1,
h2,
h3,
h4,
h5,
h6,
ul,
ol,
li {
all: revert;
}
pre {
@apply whitespace-pre-wrap rounded-lg p-2;
border: 1px solid currentColor;
}
p {
@apply mb-2;
}
/* TODO: fix markdown table */
}
@ -19,7 +32,9 @@
.btn-mini {
@apply cursor-pointer hover:shadow-md;
}
.chat-screen { max-width: 900px; }
.chat-screen {
max-width: 900px;
}
.chat-bubble-base-300 {
--tw-bg-opacity: 1;
@ -46,3 +61,7 @@
background: transparent !important;
padding: 0.5em !important;
}
.katex-display {
margin: 0 0 !important;
}

66
examples/server/webui/src/katex-gpt.js
View file

@ -1,66 +0,0 @@
import katex from 'katex';
// Adapted from https://github.com/SchneeHertz/markdown-it-katex-gpt
// MIT license
const defaultOptions = {
delimiters: [
{ left: '\\[', right: '\\]', display: true },
{ left: '\\(', right: '\\)', display: false },
],
};
export function renderLatexHTML(content, display = false) {
return katex.renderToString(content, {
throwOnError: false,
output: 'mathml',
displayMode: display,
});
}
function escapedBracketRule(options) {
return (state, silent) => {
const max = state.posMax;
const start = state.pos;
for (const { left, right, display } of options.delimiters) {
// Check if it starts with the left delimiter
if (!state.src.slice(start).startsWith(left)) continue;
// Skip the length of the left delimiter
let pos = start + left.length;
// Find the matching right delimiter
while (pos < max) {
if (state.src.slice(pos).startsWith(right)) {
break;
}
pos++;
}
// No matching right delimiter found, skip to the next match
if (pos >= max) continue;
// If not in silent mode, convert LaTeX formula to MathML
if (!silent) {
const content = state.src.slice(start + left.length, pos);
try {
const renderedContent = renderLatexHTML(content, display);
const token = state.push('html_inline', '', 0);
token.content = renderedContent;
} catch (e) {
console.error(e);
}
}
// Update position, skip the length of the right delimiter
state.pos = pos + right.length;
return true;
}
}
}
export default function (md, options = defaultOptions) {
md.inline.ruler.after('text', 'escaped_bracket', escapedBracketRule(options));
}

701
examples/server/webui/src/main.js
View file

@ -1,701 +0,0 @@
import './styles.scss';
import { createApp, defineComponent, shallowRef, computed, h } from 'vue/dist/vue.esm-bundler.js';
import MarkdownIt from 'markdown-it';
import TextLineStream from 'textlinestream';
// math formula rendering
import 'katex/dist/katex.min.css';
import markdownItKatexGpt from './katex-gpt';
import markdownItKatexNormal from '@vscode/markdown-it-katex';
// code highlighting
import hljs from './highlight-config';
import daisyuiThemes from 'daisyui/src/theming/themes';
// ponyfill for missing ReadableStream asyncIterator on Safari
import { asyncIterator } from '@sec-ant/readable-stream/ponyfill/asyncIterator';
const isDev = import.meta.env.MODE === 'development';
// types
/** @typedef {{ id: number, role: 'user' | 'assistant', content: string, timings: any }} Message */
/** @typedef {{ role: 'user' | 'assistant', content: string }} APIMessage */
/** @typedef {{ id: string, lastModified: number, messages: Array<Message> }} Conversation */
// utility functions
const isString = (x) => !!x.toLowerCase;
const isBoolean = (x) => x === true || x === false;
const isNumeric = (n) => !isString(n) && !isNaN(n) && !isBoolean(n);
const escapeAttr = (str) => str.replace(/>/g, '&gt;').replace(/"/g, '&quot;');
const copyStr = (textToCopy) => {
// Navigator clipboard api needs a secure context (https)
if (navigator.clipboard && window.isSecureContext) {
navigator.clipboard.writeText(textToCopy);
} else {
// Use the 'out of viewport hidden text area' trick
const textArea = document.createElement('textarea');
textArea.value = textToCopy;
// Move textarea out of the viewport so it's not visible
textArea.style.position = 'absolute';
textArea.style.left = '-999999px';
document.body.prepend(textArea);
textArea.select();
document.execCommand('copy');
}
};
// constants
const BASE_URL = isDev
? (localStorage.getItem('base') || 'https://localhost:8080') // for debugging
: (new URL('.', document.baseURI).href).toString().replace(/\/$/, ''); // for production
console.log({ BASE_URL });
const CONFIG_DEFAULT = {
// Note: in order not to introduce breaking changes, please keep the same data type (number, string, etc) if you want to change the default value. Do not use null or undefined for default value.
apiKey: '',
systemMessage: 'You are a helpful assistant.',
showTokensPerSecond: false,
showThoughtInProgress: false,
excludeThoughtOnReq: true,
// make sure these default values are in sync with `common.h`
samplers: 'edkypmxt',
temperature: 0.8,
dynatemp_range: 0.0,
dynatemp_exponent: 1.0,
top_k: 40,
top_p: 0.95,
min_p: 0.05,
xtc_probability: 0.0,
xtc_threshold: 0.1,
typical_p: 1.0,
repeat_last_n: 64,
repeat_penalty: 1.0,
presence_penalty: 0.0,
frequency_penalty: 0.0,
dry_multiplier: 0.0,
dry_base: 1.75,
dry_allowed_length: 2,
dry_penalty_last_n: -1,
max_tokens: -1,
custom: '', // custom json-stringified object
};
const CONFIG_INFO = {
apiKey: 'Set the API Key if you are using --api-key option for the server.',
systemMessage: 'The starting message that defines how model should behave.',
samplers: 'The order at which samplers are applied, in simplified way. Default is "dkypmxt": dry->top_k->typ_p->top_p->min_p->xtc->temperature',
temperature: 'Controls the randomness of the generated text by affecting the probability distribution of the output tokens. Higher = more random, lower = more focused.',
dynatemp_range: 'Addon for the temperature sampler. The added value to the range of dynamic temperature, which adjusts probabilities by entropy of tokens.',
dynatemp_exponent: 'Addon for the temperature sampler. Smoothes out the probability redistribution based on the most probable token.',
top_k: 'Keeps only k top tokens.',
top_p: 'Limits tokens to those that together have a cumulative probability of at least p',
min_p: 'Limits tokens based on the minimum probability for a token to be considered, relative to the probability of the most likely token.',
xtc_probability: 'XTC sampler cuts out top tokens; this parameter controls the chance of cutting tokens at all. 0 disables XTC.',
xtc_threshold: 'XTC sampler cuts out top tokens; this parameter controls the token probability that is required to cut that token.',
typical_p: 'Sorts and limits tokens based on the difference between log-probability and entropy.',
repeat_last_n: 'Last n tokens to consider for penalizing repetition',
repeat_penalty: 'Controls the repetition of token sequences in the generated text',
presence_penalty: 'Limits tokens based on whether they appear in the output or not.',
frequency_penalty: 'Limits tokens based on how often they appear in the output.',
dry_multiplier: 'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the DRY sampling multiplier.',
dry_base: 'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the DRY sampling base value.',
dry_allowed_length: 'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets the allowed length for DRY sampling.',
dry_penalty_last_n: 'DRY sampling reduces repetition in generated text even across long contexts. This parameter sets DRY penalty for the last n tokens.',
max_tokens: 'The maximum number of token per output.',
custom: '', // custom json-stringified object
};
// config keys having numeric value (i.e. temperature, top_k, top_p, etc)
const CONFIG_NUMERIC_KEYS = Object.entries(CONFIG_DEFAULT).filter(e => isNumeric(e[1])).map(e => e[0]);
// list of themes supported by daisyui
const THEMES = ['light', 'dark']
// make sure light & dark are always at the beginning
.concat(Object.keys(daisyuiThemes).filter(t => t !== 'light' && t !== 'dark'));
// markdown support
const VueMarkdown = defineComponent(
(props) => {
const md = shallowRef(new MarkdownIt({
breaks: true,
highlight: function (str, lang) { // Add highlight.js
if (lang && hljs.getLanguage(lang)) {
try {
return '<pre dir="auto"><code class="hljs">' +
hljs.highlight(str, { language: lang, ignoreIllegals: true }).value +
'</code></pre>';
} catch (__) {}
}
return '<pre dir="auto"><code class="hljs">' + md.value.utils.escapeHtml(str) + '</code></pre>';
}
}));
// support latex with double dollar sign and square brackets
md.value.use(markdownItKatexGpt, {
delimiters: [
{ left: '\\[', right: '\\]', display: true },
{ left: '\\(', right: '\\)', display: false },
{ left: '$$', right: '$$', display: false },
// do not add single dollar sign here, other wise it will confused with dollar used for money symbol
],
throwOnError: false,
});
// support latex with single dollar sign
md.value.use(markdownItKatexNormal, { throwOnError: false });
// add copy button to code blocks
const origFenchRenderer = md.value.renderer.rules.fence;
md.value.renderer.rules.fence = (tokens, idx, ...args) => {
const content = tokens[idx].content;
const origRendered = origFenchRenderer(tokens, idx, ...args);
return `<div class="relative my-4">
<div class="text-right sticky top-4 mb-2 mr-2 h-0">
<button class="badge btn-mini" onclick="copyStr(${escapeAttr(JSON.stringify(content))})">📋 Copy</button>
</div>
${origRendered}
</div>`;
};
window.copyStr = copyStr;
const content = computed(() => md.value.render(props.source));
return () => h('div', { innerHTML: content.value });
},
{ props: ['source'] }
);
// input field to be used by settings modal
const SettingsModalShortInput = defineComponent({
template: document.getElementById('settings-modal-short-input').innerHTML,
props: {
label: { type: String, required: false },
configKey: String,
configDefault: Object,
configInfo: Object,
modelValue: [Object, String, Number],
},
});
// message bubble component
const MessageBubble = defineComponent({
components: {
VueMarkdown
},
template: document.getElementById('message-bubble').innerHTML,
props: {
config: Object,
msg: Object,
isGenerating: Boolean,
showThoughtInProgress: Boolean,
editUserMsgAndRegenerate: Function,
regenerateMsg: Function,
},
data() {
return {
editingContent: null,
};
},
computed: {
timings() {
if (!this.msg.timings) return null;
return {
...this.msg.timings,
prompt_per_second: this.msg.timings.prompt_n / (this.msg.timings.prompt_ms / 1000),
predicted_per_second: this.msg.timings.predicted_n / (this.msg.timings.predicted_ms / 1000),
};
},
splitMsgContent() {
const content = this.msg.content;
if (this.msg.role !== 'assistant') {
return { content };
}
let actualContent = '';
let cot = '';
let isThinking = false;
let thinkSplit = content.split('<think>', 2);
actualContent += thinkSplit[0];
while (thinkSplit[1] !== undefined) {
// <think> tag found
thinkSplit = thinkSplit[1].split('</think>', 2);
cot += thinkSplit[0];
isThinking = true;
if (thinkSplit[1] !== undefined) {
// </think> closing tag found
isThinking = false;
thinkSplit = thinkSplit[1].split('<think>', 2);
actualContent += thinkSplit[0];
}
}
return { content: actualContent, cot, isThinking };
},
},
methods: {
copyMsg() {
copyStr(this.msg.content);
},
editMsg() {
this.editUserMsgAndRegenerate({
...this.msg,
content: this.editingContent,
});
this.editingContent = null;
},
},
});
// coversations is stored in localStorage
// format: { [convId]: { id: string, lastModified: number, messages: [...] } }
// convId is a string prefixed with 'conv-'
const StorageUtils = {
/**
* manage conversations
* @returns {Array<Conversation>}
*/
getAllConversations() {
const res = [];
for (const key in localStorage) {
if (key.startsWith('conv-')) {
res.push(JSON.parse(localStorage.getItem(key)));
}
}
res.sort((a, b) => b.lastModified - a.lastModified);
return res;
},
/**
* can return null if convId does not exist
* @param {string} convId
* @returns {Conversation | null}
*/
getOneConversation(convId) {
return JSON.parse(localStorage.getItem(convId) || 'null');
},
/**
* if convId does not exist, create one
* @param {string} convId
* @param {Message} msg
*/
appendMsg(convId, msg) {
if (msg.content === null) return;
const conv = StorageUtils.getOneConversation(convId) || {
id: convId,
lastModified: Date.now(),
messages: [],
};
conv.messages.push(msg);
conv.lastModified = Date.now();
localStorage.setItem(convId, JSON.stringify(conv));
},
/**
* Get new conversation id
* @returns {string}
*/
getNewConvId() {
return `conv-${Date.now()}`;
},
/**
* remove conversation by id
* @param {string} convId
*/
remove(convId) {
localStorage.removeItem(convId);
},
/**
* remove all conversations
* @param {string} convId
*/
filterAndKeepMsgs(convId, predicate) {
const conv = StorageUtils.getOneConversation(convId);
if (!conv) return;
conv.messages = conv.messages.filter(predicate);
conv.lastModified = Date.now();
localStorage.setItem(convId, JSON.stringify(conv));
},
/**
* remove last message from conversation
* @param {string} convId
* @returns {Message | undefined}
*/
popMsg(convId) {
const conv = StorageUtils.getOneConversation(convId);
if (!conv) return;
const msg = conv.messages.pop();
conv.lastModified = Date.now();
if (conv.messages.length === 0) {
StorageUtils.remove(convId);
} else {
localStorage.setItem(convId, JSON.stringify(conv));
}
return msg;
},
// manage config
getConfig() {
const savedVal = JSON.parse(localStorage.getItem('config') || '{}');
// to prevent breaking changes in the future, we always provide default value for missing keys
return {
...CONFIG_DEFAULT,
...savedVal,
};
},
setConfig(config) {
localStorage.setItem('config', JSON.stringify(config));
},
getTheme() {
return localStorage.getItem('theme') || 'auto';
},
setTheme(theme) {
if (theme === 'auto') {
localStorage.removeItem('theme');
} else {
localStorage.setItem('theme', theme);
}
},
};
// scroll to bottom of chat messages
// if requiresNearBottom is true, only auto-scroll if user is near bottom
const chatScrollToBottom = (requiresNearBottom) => {
const msgListElem = document.getElementById('messages-list');
const spaceToBottom = msgListElem.scrollHeight - msgListElem.scrollTop - msgListElem.clientHeight;
if (!requiresNearBottom || (spaceToBottom < 100)) {
setTimeout(() => msgListElem.scrollTo({ top: msgListElem.scrollHeight }), 1);
}
};
// wrapper for SSE
async function* sendSSEPostRequest(url, fetchOptions) {
const res = await fetch(url, fetchOptions);
const lines = res.body
.pipeThrough(new TextDecoderStream())
.pipeThrough(new TextLineStream());
for await (const line of asyncIterator(lines)) {
if (isDev) console.log({line});
if (line.startsWith('data:') && !line.endsWith('[DONE]')) {
const data = JSON.parse(line.slice(5));
yield data;
} else if (line.startsWith('error:')) {
const data = JSON.parse(line.slice(6));
throw new Error(data.message || 'Unknown error');
}
}
};
const mainApp = createApp({
components: {
VueMarkdown,
SettingsModalShortInput,
MessageBubble,
},
data() {
return {
conversations: StorageUtils.getAllConversations(),
/** @type {Array<Message>} */
messages: [],
viewingConvId: StorageUtils.getNewConvId(),
inputMsg: '',
isGenerating: false,
/** @type {Array<Message> | null} */
pendingMsg: null, // the on-going message from assistant
stopGeneration: () => {},
selectedTheme: StorageUtils.getTheme(),
config: StorageUtils.getConfig(),
showConfigDialog: false,
// const
themes: THEMES,
/** @type {CONFIG_DEFAULT} */
configDefault: {...CONFIG_DEFAULT},
configInfo: {...CONFIG_INFO},
isDev,
}
},
computed: {},
mounted() {
document.getElementById('app').classList.remove('opacity-0'); // show app
// scroll to the bottom when the pending message height is updated
const pendingMsgElem = document.getElementById('pending-msg');
const resizeObserver = new ResizeObserver(() => {
if (this.isGenerating) chatScrollToBottom(true);
});
resizeObserver.observe(pendingMsgElem);
this.setSelectedTheme(this.selectedTheme);
},
watch: {
viewingConvId: function(val, oldVal) {
if (val != oldVal) {
this.fetchMessages();
chatScrollToBottom();
this.hideSidebar();
}
}
},
methods: {
hideSidebar() {
document.getElementById('toggle-drawer').checked = false;
},
setSelectedTheme(theme) {
this.selectedTheme = theme;
document.body.setAttribute('data-theme', theme);
document.body.setAttribute('data-color-scheme', daisyuiThemes[theme]?.['color-scheme'] ?? 'auto');
StorageUtils.setTheme(theme);
},
newConversation() {
if (this.isGenerating) return;
this.viewingConvId = StorageUtils.getNewConvId();
},
setViewingConv(convId) {
if (this.isGenerating) return;
this.viewingConvId = convId;
},
deleteConv(convId) {
if (this.isGenerating) return;
if (window.confirm('Are you sure to delete this conversation?')) {
StorageUtils.remove(convId);
if (this.viewingConvId === convId) {
this.viewingConvId = StorageUtils.getNewConvId();
}
this.fetchConversation();
this.fetchMessages();
}
},
downloadConv(convId) {
const conversation = StorageUtils.getOneConversation(convId);
if (!conversation) {
alert('Conversation not found.');
return;
}
const conversationJson = JSON.stringify(conversation, null, 2);
const blob = new Blob([conversationJson], { type: 'application/json' });
const url = URL.createObjectURL(blob);
const a = document.createElement('a');
a.href = url;
a.download = `conversation_${convId}.json`;
document.body.appendChild(a);
a.click();
document.body.removeChild(a);
URL.revokeObjectURL(url);
},
async sendMessage() {
if (!this.inputMsg) return;
const currConvId = this.viewingConvId;
StorageUtils.appendMsg(currConvId, {
id: Date.now(),
role: 'user',
content: this.inputMsg,
});
this.fetchConversation();
this.fetchMessages();
this.inputMsg = '';
this.generateMessage(currConvId);
chatScrollToBottom();
},
async generateMessage(currConvId) {
if (this.isGenerating) return;
this.pendingMsg = { id: Date.now()+1, role: 'assistant', content: null };
this.isGenerating = true;
try {
/** @type {CONFIG_DEFAULT} */
const config = this.config;
const abortController = new AbortController();
this.stopGeneration = () => abortController.abort();
/** @type {Array<APIMessage>} */
let messages = [
{ role: 'system', content: config.systemMessage },
...normalizeMsgsForAPI(this.messages),
];
if (config.excludeThoughtOnReq) {
messages = filterThoughtFromMsgs(messages);
}
if (isDev) console.log({messages});
const params = {
messages,
stream: true,
cache_prompt: true,
samplers: config.samplers,
temperature: config.temperature,
dynatemp_range: config.dynatemp_range,
dynatemp_exponent: config.dynatemp_exponent,
top_k: config.top_k,
top_p: config.top_p,
min_p: config.min_p,
typical_p: config.typical_p,
xtc_probability: config.xtc_probability,
xtc_threshold: config.xtc_threshold,
repeat_last_n: config.repeat_last_n,
repeat_penalty: config.repeat_penalty,
presence_penalty: config.presence_penalty,
frequency_penalty: config.frequency_penalty,
dry_multiplier: config.dry_multiplier,
dry_base: config.dry_base,
dry_allowed_length: config.dry_allowed_length,
dry_penalty_last_n: config.dry_penalty_last_n,
max_tokens: config.max_tokens,
timings_per_token: !!config.showTokensPerSecond,
...(config.custom.length ? JSON.parse(config.custom) : {}),
};
const chunks = sendSSEPostRequest(`${BASE_URL}/v1/chat/completions`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
...(config.apiKey ? {'Authorization': `Bearer ${config.apiKey}`} : {})
},
body: JSON.stringify(params),
signal: abortController.signal,
});
for await (const chunk of chunks) {
const stop = chunk.stop;
const addedContent = chunk.choices[0].delta.content;
const lastContent = this.pendingMsg.content || '';
if (addedContent) {
this.pendingMsg = {
id: this.pendingMsg.id,
role: 'assistant',
content: lastContent + addedContent,
};
}
const timings = chunk.timings;
if (timings && config.showTokensPerSecond) {
// only extract what's really needed, to save some space
this.pendingMsg.timings = {
prompt_n: timings.prompt_n,
prompt_ms: timings.prompt_ms,
predicted_n: timings.predicted_n,
predicted_ms: timings.predicted_ms,
};
}
}
StorageUtils.appendMsg(currConvId, this.pendingMsg);
this.fetchConversation();
this.fetchMessages();
setTimeout(() => document.getElementById('msg-input').focus(), 1);
} catch (error) {
if (error.name === 'AbortError') {
// user stopped the generation via stopGeneration() function
StorageUtils.appendMsg(currConvId, this.pendingMsg);
this.fetchConversation();
this.fetchMessages();
} else {
console.error(error);
alert(error);
// pop last user message
const lastUserMsg = StorageUtils.popMsg(currConvId);
this.inputMsg = lastUserMsg ? lastUserMsg.content : '';
}
}
this.pendingMsg = null;
this.isGenerating = false;
this.stopGeneration = () => {};
this.fetchMessages();
chatScrollToBottom();
},
// message actions
regenerateMsg(msg) {
if (this.isGenerating) return;
// TODO: somehow keep old history (like how ChatGPT has different "tree"). This can be done by adding "sub-conversations" with "subconv-" prefix, and new message will have a list of subconvIds
const currConvId = this.viewingConvId;
StorageUtils.filterAndKeepMsgs(currConvId, (m) => m.id < msg.id);
this.fetchConversation();
this.fetchMessages();
this.generateMessage(currConvId);
},
editUserMsgAndRegenerate(msg) {
if (this.isGenerating) return;
const currConvId = this.viewingConvId;
const newContent = msg.content;
StorageUtils.filterAndKeepMsgs(currConvId, (m) => m.id < msg.id);
StorageUtils.appendMsg(currConvId, {
id: Date.now(),
role: 'user',
content: newContent,
});
this.fetchConversation();
this.fetchMessages();
this.generateMessage(currConvId);
},
// settings dialog methods
closeAndSaveConfigDialog() {
try {
if (this.config.custom.length) JSON.parse(this.config.custom);
} catch (error) {
alert('Invalid JSON for custom config. Please either fix it or leave it empty.');
return;
}
for (const key of CONFIG_NUMERIC_KEYS) {
if (isNaN(this.config[key]) || this.config[key].toString().trim().length === 0) {
alert(`Invalid number for ${key} (expected an integer or a float)`);
return;
}
this.config[key] = parseFloat(this.config[key]);
}
this.showConfigDialog = false;
StorageUtils.setConfig(this.config);
},
closeAndDiscardConfigDialog() {
this.showConfigDialog = false;
this.config = StorageUtils.getConfig();
},
resetConfigDialog() {
if (window.confirm('Are you sure to reset all settings?')) {
this.config = {...CONFIG_DEFAULT};
}
},
// sync state functions
fetchConversation() {
this.conversations = StorageUtils.getAllConversations();
},
fetchMessages() {
this.messages = StorageUtils.getOneConversation(this.viewingConvId)?.messages ?? [];
},
// debug functions
async debugImportDemoConv() {
const res = await fetch('/demo-conversation.json');
const demoConv = await res.json();
StorageUtils.remove(demoConv.id);
for (const msg of demoConv.messages) {
StorageUtils.appendMsg(demoConv.id, msg);
}
this.fetchConversation();
}
},
});
mainApp.config.errorHandler = alert;
try {
mainApp.mount('#app');
} catch (err) {
console.error(err);
document.getElementById('app').innerHTML = `<div style="margin:2em auto">
Failed to start app. Please try clearing localStorage and try again.<br/>
<br/>
<button class="btn" onClick="localStorage.clear(); window.location.reload();">Clear localStorage</button>
</div>`;
}
/**
* filter out redundant fields upon sending to API
* @param {Array<APIMessage>} messages
* @returns {Array<APIMessage>}
*/
function normalizeMsgsForAPI(messages) {
return messages.map((msg) => {
return {
role: msg.role,
content: msg.content,
};
});
}
/**
* recommended for DeepsSeek-R1, filter out content between <think> and </think> tags
* @param {Array<APIMessage>} messages
* @returns {Array<APIMessage>}
*/
function filterThoughtFromMsgs(messages) {
return messages.map((msg) => {
return {
role: msg.role,
content: msg.role === 'assistant'
? msg.content.split('</think>').at(-1).trim()
: msg.content,
};
});
}

10
examples/server/webui/src/main.tsx Normal file
View file

@ -0,0 +1,10 @@
import { StrictMode } from 'react';
import { createRoot } from 'react-dom/client';
import './index.scss';
import App from './App.tsx';
createRoot(document.getElementById('root')!).render(
<StrictMode>
<App />
</StrictMode>
);

304
examples/server/webui/src/utils/app.context.tsx Normal file
View file

@ -0,0 +1,304 @@
import React, { createContext, useContext, useEffect, useState } from 'react';
import { APIMessage, Conversation, Message, PendingMessage } from './types';
import StorageUtils from './storage';
import {
filterThoughtFromMsgs,
normalizeMsgsForAPI,
getSSEStreamAsync,
} from './misc';
import { BASE_URL, CONFIG_DEFAULT, isDev } from '../Config';
import { matchPath, useLocation } from 'react-router';
interface AppContextValue {
viewingConversation: Conversation | null;
pendingMessages: Record<Conversation['id'], PendingMessage>;
isGenerating: (convId: string) => boolean;
sendMessage: (
convId: string,
content: string,
onChunk?: CallbackGeneratedChunk
) => Promise<boolean>;
stopGenerating: (convId: string) => void;
replaceMessageAndGenerate: (
convId: string,
origMsgId: Message['id'],
content?: string,
onChunk?: CallbackGeneratedChunk
) => Promise<void>;
config: typeof CONFIG_DEFAULT;
saveConfig: (config: typeof CONFIG_DEFAULT) => void;
}
// for now, this callback is only used for scrolling to the bottom of the chat
type CallbackGeneratedChunk = () => void;
// eslint-disable-next-line @typescript-eslint/no-explicit-any
const AppContext = createContext<AppContextValue>({} as any);
export const AppContextProvider = ({
children,
}: {
children: React.ReactElement;
}) => {
const { pathname } = useLocation();
const params = matchPath('/chat/:convId', pathname);
const convId = params?.params?.convId;
const [viewingConversation, setViewingConversation] =
useState<Conversation | null>(null);
const [pendingMessages, setPendingMessages] = useState<
Record<Conversation['id'], PendingMessage>
>({});
const [aborts, setAborts] = useState<
Record<Conversation['id'], AbortController>
>({});
const [config, setConfig] = useState(StorageUtils.getConfig());
useEffect(() => {
const handleConversationChange = (changedConvId: string) => {
if (changedConvId !== convId) return;
setViewingConversation(StorageUtils.getOneConversation(convId));
};
StorageUtils.onConversationChanged(handleConversationChange);
setViewingConversation(StorageUtils.getOneConversation(convId ?? ''));
return () => {
StorageUtils.offConversationChanged(handleConversationChange);
};
}, [convId]);
const setPending = (convId: string, pendingMsg: PendingMessage | null) => {
// if pendingMsg is null, remove the key from the object
if (!pendingMsg) {
setPendingMessages((prev) => {
const newState = { ...prev };
delete newState[convId];
return newState;
});
} else {
setPendingMessages((prev) => ({ ...prev, [convId]: pendingMsg }));
}
};
const setAbort = (convId: string, controller: AbortController | null) => {
if (!controller) {
setAborts((prev) => {
const newState = { ...prev };
delete newState[convId];
return newState;
});
} else {
setAborts((prev) => ({ ...prev, [convId]: controller }));
}
};
////////////////////////////////////////////////////////////////////////
// public functions
const isGenerating = (convId: string) => !!pendingMessages[convId];
const generateMessage = async (
convId: string,
onChunk?: CallbackGeneratedChunk
) => {
if (isGenerating(convId)) return;
const config = StorageUtils.getConfig();
const currConversation = StorageUtils.getOneConversation(convId);
if (!currConversation) {
throw new Error('Current conversation is not found');
}
const abortController = new AbortController();
setAbort(convId, abortController);
let pendingMsg: PendingMessage = {
id: Date.now() + 1,
role: 'assistant',
content: null,
};
setPending(convId, pendingMsg);
try {
// prepare messages for API
let messages: APIMessage[] = [
...(config.systemMessage.length === 0
? []
: [{ role: 'system', content: config.systemMessage } as APIMessage]),
...normalizeMsgsForAPI(currConversation?.messages ?? []),
];
if (config.excludeThoughtOnReq) {
messages = filterThoughtFromMsgs(messages);
}
if (isDev) console.log({ messages });
// prepare params
const params = {
messages,
stream: true,
cache_prompt: true,
samplers: config.samplers,
temperature: config.temperature,
dynatemp_range: config.dynatemp_range,
dynatemp_exponent: config.dynatemp_exponent,
top_k: config.top_k,
top_p: config.top_p,
min_p: config.min_p,
typical_p: config.typical_p,
xtc_probability: config.xtc_probability,
xtc_threshold: config.xtc_threshold,
repeat_last_n: config.repeat_last_n,
repeat_penalty: config.repeat_penalty,
presence_penalty: config.presence_penalty,
frequency_penalty: config.frequency_penalty,
dry_multiplier: config.dry_multiplier,
dry_base: config.dry_base,
dry_allowed_length: config.dry_allowed_length,
dry_penalty_last_n: config.dry_penalty_last_n,
max_tokens: config.max_tokens,
timings_per_token: !!config.showTokensPerSecond,
...(config.custom.length ? JSON.parse(config.custom) : {}),
};
// send request
const fetchResponse = await fetch(`${BASE_URL}/v1/chat/completions`, {
method: 'POST',
headers: {
'Content-Type': 'application/json',
...(config.apiKey
? { Authorization: `Bearer ${config.apiKey}` }
: {}),
},
body: JSON.stringify(params),
signal: abortController.signal,
});
if (fetchResponse.status !== 200) {
const body = await fetchResponse.json();
throw new Error(body?.error?.message || 'Unknown error');
}
const chunks = getSSEStreamAsync(fetchResponse);
for await (const chunk of chunks) {
// const stop = chunk.stop;
if (chunk.error) {
throw new Error(chunk.error?.message || 'Unknown error');
}
const addedContent = chunk.choices[0].delta.content;
const lastContent = pendingMsg.content || '';
if (addedContent) {
pendingMsg = {
id: pendingMsg.id,
role: 'assistant',
content: lastContent + addedContent,
};
}
const timings = chunk.timings;
if (timings && config.showTokensPerSecond) {
// only extract what's really needed, to save some space
pendingMsg.timings = {
prompt_n: timings.prompt_n,
prompt_ms: timings.prompt_ms,
predicted_n: timings.predicted_n,
predicted_ms: timings.predicted_ms,
};
}
setPending(convId, pendingMsg);
onChunk?.();
}
} catch (err) {
setPending(convId, null);
if ((err as Error).name === 'AbortError') {
// user stopped the generation via stopGeneration() function
// we can safely ignore this error
} else {
console.error(err);
// eslint-disable-next-line @typescript-eslint/no-explicit-any
alert((err as any)?.message ?? 'Unknown error');
throw err; // rethrow
}
}
if (pendingMsg.content) {
StorageUtils.appendMsg(currConversation.id, {
id: pendingMsg.id,
content: pendingMsg.content,
role: pendingMsg.role,
timings: pendingMsg.timings,
});
}
setPending(convId, null);
onChunk?.(); // trigger scroll to bottom
};
const sendMessage = async (
convId: string,
content: string,
onChunk?: CallbackGeneratedChunk
): Promise<boolean> => {
if (isGenerating(convId) || content.trim().length === 0) return false;
StorageUtils.appendMsg(convId, {
id: Date.now(),
role: 'user',
content,
});
try {
await generateMessage(convId, onChunk);
return true;
} catch (_) {
// rollback
StorageUtils.popMsg(convId);
}
return false;
};
const stopGenerating = (convId: string) => {
setPending(convId, null);
aborts[convId]?.abort();
};
// if content is undefined, we remove last assistant message
const replaceMessageAndGenerate = async (
convId: string,
origMsgId: Message['id'],
content?: string,
onChunk?: CallbackGeneratedChunk
) => {
if (isGenerating(convId)) return;
StorageUtils.filterAndKeepMsgs(convId, (msg) => msg.id < origMsgId);
if (content) {
StorageUtils.appendMsg(convId, {
id: Date.now(),
role: 'user',
content,
});
}
await generateMessage(convId, onChunk);
};
const saveConfig = (config: typeof CONFIG_DEFAULT) => {
StorageUtils.setConfig(config);
setConfig(config);
};
return (
<AppContext.Provider
value={{
isGenerating,
viewingConversation,
pendingMessages,
sendMessage,
stopGenerating,
replaceMessageAndGenerate,
config,
saveConfig,
}}
>
{children}
</AppContext.Provider>
);
};
export const useAppContext = () => useContext(AppContext);

87
examples/server/webui/src/utils/misc.ts Normal file
View file

@ -0,0 +1,87 @@
// @ts-expect-error this package does not have typing
import TextLineStream from 'textlinestream';
import { APIMessage, Message } from './types';
// ponyfill for missing ReadableStream asyncIterator on Safari
import { asyncIterator } from '@sec-ant/readable-stream/ponyfill/asyncIterator';
import { isDev } from '../Config';
// eslint-disable-next-line @typescript-eslint/no-explicit-any
export const isString = (x: any) => !!x.toLowerCase;
// eslint-disable-next-line @typescript-eslint/no-explicit-any
export const isBoolean = (x: any) => x === true || x === false;
// eslint-disable-next-line @typescript-eslint/no-explicit-any
export const isNumeric = (n: any) => !isString(n) && !isNaN(n) && !isBoolean(n);
export const escapeAttr = (str: string) =>
str.replace(/>/g, '&gt;').replace(/"/g, '&quot;');
// wrapper for SSE
export async function* getSSEStreamAsync(fetchResponse: Response) {
if (!fetchResponse.body) throw new Error('Response body is empty');
const lines: ReadableStream<string> = fetchResponse.body
.pipeThrough(new TextDecoderStream())
.pipeThrough(new TextLineStream());
// @ts-expect-error asyncIterator complains about type, but it should work
for await (const line of asyncIterator(lines)) {
if (isDev) console.log({ line });
if (line.startsWith('data:') && !line.endsWith('[DONE]')) {
const data = JSON.parse(line.slice(5));
yield data;
} else if (line.startsWith('error:')) {
const data = JSON.parse(line.slice(6));
throw new Error(data.message || 'Unknown error');
}
}
}
// copy text to clipboard
export const copyStr = (textToCopy: string) => {
// Navigator clipboard api needs a secure context (https)
if (navigator.clipboard && window.isSecureContext) {
navigator.clipboard.writeText(textToCopy);
} else {
// Use the 'out of viewport hidden text area' trick
const textArea = document.createElement('textarea');
textArea.value = textToCopy;
// Move textarea out of the viewport so it's not visible
textArea.style.position = 'absolute';
textArea.style.left = '-999999px';
document.body.prepend(textArea);
textArea.select();
document.execCommand('copy');
}
};
/**
* filter out redundant fields upon sending to API
*/
export function normalizeMsgsForAPI(messages: Message[]) {
return messages.map((msg) => {
return {
role: msg.role,
content: msg.content,
};
}) as APIMessage[];
}
/**
* recommended for DeepsSeek-R1, filter out content between <think> and </think> tags
*/
export function filterThoughtFromMsgs(messages: APIMessage[]) {
return messages.map((msg) => {
return {
role: msg.role,
content:
msg.role === 'assistant'
? msg.content.split('</think>').at(-1)!.trim()
: msg.content,
} as APIMessage;
});
}
export function classNames(classes: Record<string, boolean>): string {
return Object.entries(classes)
.filter(([_, value]) => value)
.map(([key, _]) => key)
.join(' ');
}

138
examples/server/webui/src/utils/storage.ts Normal file
View file

@ -0,0 +1,138 @@
// coversations is stored in localStorage
// format: { [convId]: { id: string, lastModified: number, messages: [...] } }
import { CONFIG_DEFAULT } from '../Config';
import { Conversation, Message } from './types';
const event = new EventTarget();
type CallbackConversationChanged = (convId: string) => void;
let onConversationChangedHandlers: [
CallbackConversationChanged,
EventListener,
][] = [];
const dispatchConversationChange = (convId: string) => {
event.dispatchEvent(
new CustomEvent('conversationChange', { detail: { convId } })
);
};
// convId is a string prefixed with 'conv-'
const StorageUtils = {
/**
* manage conversations
*/
getAllConversations(): Conversation[] {
const res = [];
for (const key in localStorage) {
if (key.startsWith('conv-')) {
res.push(JSON.parse(localStorage.getItem(key) ?? '{}'));
}
}
res.sort((a, b) => b.lastModified - a.lastModified);
return res;
},
/**
* can return null if convId does not exist
*/
getOneConversation(convId: string): Conversation | null {
return JSON.parse(localStorage.getItem(convId) || 'null');
},
/**
* if convId does not exist, create one
*/
appendMsg(convId: string, msg: Message): void {
if (msg.content === null) return;
const conv = StorageUtils.getOneConversation(convId) || {
id: convId,
lastModified: Date.now(),
messages: [],
};
conv.messages.push(msg);
conv.lastModified = Date.now();
localStorage.setItem(convId, JSON.stringify(conv));
dispatchConversationChange(convId);
},
/**
* Get new conversation id
*/
getNewConvId(): string {
return `conv-${Date.now()}`;
},
/**
* remove conversation by id
*/
remove(convId: string): void {
localStorage.removeItem(convId);
dispatchConversationChange(convId);
},
/**
* remove all conversations
*/
filterAndKeepMsgs(
convId: string,
predicate: (msg: Message) => boolean
): void {
const conv = StorageUtils.getOneConversation(convId);
if (!conv) return;
conv.messages = conv.messages.filter(predicate);
conv.lastModified = Date.now();
localStorage.setItem(convId, JSON.stringify(conv));
dispatchConversationChange(convId);
},
/**
* remove last message from conversation
*/
popMsg(convId: string): Message | undefined {
const conv = StorageUtils.getOneConversation(convId);
if (!conv) return;
const msg = conv.messages.pop();
conv.lastModified = Date.now();
if (conv.messages.length === 0) {
StorageUtils.remove(convId);
} else {
localStorage.setItem(convId, JSON.stringify(conv));
}
dispatchConversationChange(convId);
return msg;
},
// event listeners
onConversationChanged(callback: CallbackConversationChanged) {
const fn = (e: Event) => callback((e as CustomEvent).detail.convId);
onConversationChangedHandlers.push([callback, fn]);
event.addEventListener('conversationChange', fn);
},
offConversationChanged(callback: CallbackConversationChanged) {
const fn = onConversationChangedHandlers.find(([cb, _]) => cb === callback);
if (fn) {
event.removeEventListener('conversationChange', fn[1]);
}
onConversationChangedHandlers = [];
},
// manage config
getConfig(): typeof CONFIG_DEFAULT {
const savedVal = JSON.parse(localStorage.getItem('config') || '{}');
// to prevent breaking changes in the future, we always provide default value for missing keys
return {
...CONFIG_DEFAULT,
...savedVal,
};
},
setConfig(config: typeof CONFIG_DEFAULT) {
localStorage.setItem('config', JSON.stringify(config));
},
getTheme(): string {
return localStorage.getItem('theme') || 'auto';
},
setTheme(theme: string) {
if (theme === 'auto') {
localStorage.removeItem('theme');
} else {
localStorage.setItem('theme', theme);
}
},
};
export default StorageUtils;

25
examples/server/webui/src/utils/types.ts Normal file
View file

@ -0,0 +1,25 @@
export interface TimingReport {
prompt_n: number;
prompt_ms: number;
predicted_n: number;
predicted_ms: number;
}
export interface Message {
id: number;
role: 'user' | 'assistant' | 'system';
content: string;
timings?: TimingReport;
}
export type APIMessage = Pick<Message, 'role' | 'content'>;
export interface Conversation {
id: string; // format: `conv-{timestamp}`
lastModified: number; // timestamp from Date.now()
messages: Message[];
}
export type PendingMessage = Omit<Message, 'content'> & {
content: string | null;
};

1
examples/server/webui/src/vite-env.d.ts vendored Normal file
View file

@ -0,0 +1 @@
/// <reference types="vite/client" />

26
examples/server/webui/tsconfig.app.json Normal file
View file

@ -0,0 +1,26 @@
{
"compilerOptions": {
"tsBuildInfoFile": "./node_modules/.tmp/tsconfig.app.tsbuildinfo",
"target": "ES2021",
"useDefineForClassFields": true,
"lib": ["ES2021", "DOM", "DOM.Iterable"],
"module": "ESNext",
"skipLibCheck": true,
/* Bundler mode */
"moduleResolution": "bundler",
"allowImportingTsExtensions": true,
"isolatedModules": true,
"moduleDetection": "force",
"noEmit": true,
"jsx": "react-jsx",
/* Linting */
"strict": true,
"noUnusedLocals": true,
"noUnusedParameters": true,
"noFallthroughCasesInSwitch": true,
"noUncheckedSideEffectImports": true
},
"include": ["src"]
}

7
examples/server/webui/tsconfig.json Normal file
View file

@ -0,0 +1,7 @@
{
"files": [],
"references": [
{ "path": "./tsconfig.app.json" },
{ "path": "./tsconfig.node.json" }
]
}

24
examples/server/webui/tsconfig.node.json Normal file
View file

@ -0,0 +1,24 @@
{
"compilerOptions": {
"tsBuildInfoFile": "./node_modules/.tmp/tsconfig.node.tsbuildinfo",
"target": "ES2022",
"lib": ["ES2023"],
"module": "ESNext",
"skipLibCheck": true,
/* Bundler mode */
"moduleResolution": "bundler",
"allowImportingTsExtensions": true,
"isolatedModules": true,
"moduleDetection": "force",
"noEmit": true,
/* Linting */
"strict": true,
"noUnusedLocals": true,
"noUnusedParameters": true,
"noFallthroughCasesInSwitch": true,
"noUncheckedSideEffectImports": true
},
"include": ["vite.config.ts"]
}

49
examples/server/webui/vite.config.js → examples/server/webui/vite.config.ts
View file

@ -1,8 +1,11 @@
import { defineConfig, PluginOption } from 'vite';
import react from '@vitejs/plugin-react';
import { viteSingleFile } from 'vite-plugin-singlefile';
import path from 'path';
import fs from 'fs';
import zlib from 'zlib';
import path from 'node:path';
import fs from 'node:fs';
import zlib from 'node:zlib';
/* eslint-disable */
const MAX_BUNDLE_SIZE = 1.5 * 1024 * 1024; // only increase when absolutely necessary
@ -15,20 +18,26 @@ const GUIDE_FOR_FRONTEND = `
-->
`.trim();
const FRONTEND_PLUGINS = [react()];
const BUILD_PLUGINS = [
...FRONTEND_PLUGINS,
viteSingleFile(),
(function llamaCppPlugin() {
let config;
let config: any;
return {
name: 'llamacpp:build',
apply: 'build',
async configResolved(_config) {
async configResolved(_config: any) {
config = _config;
},
writeBundle() {
const outputIndexHtml = path.join(config.build.outDir, 'index.html');
const content = GUIDE_FOR_FRONTEND + '\n' + fs.readFileSync(outputIndexHtml, 'utf-8');
const compressed = zlib.gzipSync(Buffer.from(content, 'utf-8'), { level: 9 });
const content =
GUIDE_FOR_FRONTEND + '\n' + fs.readFileSync(outputIndexHtml, 'utf-8');
const compressed = zlib.gzipSync(Buffer.from(content, 'utf-8'), {
level: 9,
});
// because gzip header contains machine-specific info, we must remove these data from the header
// timestamp
@ -42,18 +51,26 @@ const BUILD_PLUGINS = [
if (compressed.byteLength > MAX_BUNDLE_SIZE) {
throw new Error(
`Bundle size is too large (${Math.ceil(compressed.byteLength / 1024)} KB).\n` +
`Please reduce the size of the frontend or increase MAX_BUNDLE_SIZE in vite.config.js.\n`,
`Please reduce the size of the frontend or increase MAX_BUNDLE_SIZE in vite.config.js.\n`
);
}
const targetOutputFile = path.join(config.build.outDir, '../../public/index.html.gz');
const targetOutputFile = path.join(
config.build.outDir,
'../../public/index.html.gz'
);
fs.writeFileSync(targetOutputFile, compressed);
}
}
},
} satisfies PluginOption;
})(),
];
/** @type {import('vite').UserConfig} */
export default {
plugins: process.env.ANALYZE ? [] : BUILD_PLUGINS,
};
export default defineConfig({
// @ts-ignore
plugins: process.env.ANALYZE ? FRONTEND_PLUGINS : BUILD_PLUGINS,
server: {
proxy: {
'/v1': 'http://localhost:8080',
},
},
});

11
examples/simple-cmake-pkg/CMakeLists.txt
View file

@ -1,11 +0,0 @@
cmake_minimum_required(VERSION 3.12)
project(llama-simple-cmake-pkg)
set(TARGET llama-simple-cmake-pkg)
find_package(Llama REQUIRED)
add_executable(${TARGET} ${CMAKE_CURRENT_LIST_DIR}/../simple/simple.cpp)
install(TARGETS ${TARGET} RUNTIME)
target_link_libraries(${TARGET} PRIVATE llama ggml::all ${CMAKE_THREAD_LIBS_INIT})
target_compile_features(${TARGET} PRIVATE cxx_std_17)

34
examples/simple-cmake-pkg/README.md
View file

@ -1,34 +0,0 @@
# llama.cpp/example/simple-cmake-pkg
This program builds [simple](../simple) using a relocatable CMake package. It serves as an example of using the `find_package()` CMake command to conveniently include [llama.cpp](https://github.com/ggerganov/llama.cpp) in projects which live outside of the source tree.
## Building
Because this example is "outside of the source tree", it is important to first build/install llama.cpp using CMake. An example is provided here, but please see the [llama.cpp build instructions](../..) for more detailed build instructions.
### Considerations
When hardware acceleration libraries are used (e.g. CUDA, Metal, Vulkan, etc.), the appropriate dependencies will be searched for automatically. So, for example, when finding a package
### Build llama.cpp and install to llama.cpp/inst
```sh
git clone https://github.com/ggerganov/llama.cpp
cd llama.cpp
cmake -S . -B build
cmake --build build
cmake --install build --prefix inst
### Build simple-cmake-pkg
```sh
cd examples/simple-cmake-pkg
cmake -S . -B build -DCMAKE_PREFIX_PATH=../../inst/lib/cmake
cmake --build build
```
### Run simple-cmake-pkg
```sh
./build/llama-simple-cmake-pkg -m ./models/llama-7b-v2/ggml-model-f16.gguf "Hello my name is"
```

147
ggml/cmake/ggml-config.cmake.in
View file

@ -1,147 +0,0 @@
@GGML_VARIABLES_EXPANDED@
@PACKAGE_INIT@
set_and_check(GGML_INCLUDE_DIR "@PACKAGE_GGML_INCLUDE_INSTALL_DIR@")
set_and_check(GGML_LIB_DIR "@PACKAGE_GGML_LIB_INSTALL_DIR@")
set_and_check(GGML_BIN_DIR "@PACKAGE_GGML_BIN_INSTALL_DIR@")
find_package(Threads REQUIRED)
find_library(GGML_LIBRARY ggml
REQUIRED
HINTS ${GGML_LIB_DIR}
NO_CMAKE_FIND_ROOT_PATH)
add_library(ggml::ggml UNKNOWN IMPORTED)
set_target_properties(ggml::ggml
PROPERTIES
IMPORTED_LOCATION "${GGML_LIBRARY}")
find_library(GGML_BASE_LIBRARY ggml-base
REQUIRED
HINTS ${GGML_LIB_DIR}
NO_CMAKE_FIND_ROOT_PATH)
add_library(ggml::ggml-base UNKNOWN IMPORTED)
set_target_properties(ggml::ggml-base
PROPERTIES
IMPORTED_LOCATION "${GGML_BASE_LIBRARY}")
if (NOT GGML_SHARED_LIB)
if (APPLE AND GGML_ACCELERATE)
find_library(ACCELERATE_FRAMEWORK Accelerate REQUIRED)
list(APPEND GGML_CPU_INTERFACE_LINK_LIBRARIES ${ACCELERATE_FRAMEWORK})
endif()
if (GGML_OPENMP)
find_package(OpenMP REQUIRED)
list(APPEND GGML_CPU_INTERFACE_LINK_LIBRARIES OpenMP::OpenMP_C OpenMP::OpenMP_CXX)
endif()
if (GGML_CPU_HBM)
find_library(memkind memkind REQUIRED)
list(APPEND GGML_CPU_INTERFACE_LINK_LIBRARIES memkind)
endif()
if (GGML_BLAS)
find_package(BLAS REQUIRED)
list(APPEND GGML_CPU_INTERFACE_LINK_LIBRARIES ${BLAS_LIBRARIES})
list(APPEND GGML_CPU_INTERFACE_LINK_OPTIONS ${BLAS_LINKER_FLAGS})
endif()
if (GGML_CUDA)
find_package(CUDAToolkit REQUIRED)
endif()
if (GGML_METAL)
find_library(FOUNDATION_LIBRARY Foundation REQUIRED)
find_library(METAL_FRAMEWORK Metal REQUIRED)
find_library(METALKIT_FRAMEWORK MetalKit REQUIRED)
list(APPEND GGML_METAL_INTERFACE_LINK_LIBRARIES
${FOUNDATION_LIBRARY} ${METAL_FRAMEWORK} ${METALKIT_FRAMEWORK})
endif()
if (GGML_VULKAN)
find_package(Vulkan REQUIRED)
list(APPEND GGML_VULKAN_INTERFACE_LINK_LIBRARIES Vulkan::Vulkan)
endif()
if (GGML_HIP)
find_package(hip REQUIRED)
find_package(hipblas REQUIRED)
find_package(rocblas REQUIRED)
list(APPEND GGML_HIP_INTERFACE_LINK_LIBRARIES hip::host roc::rocblas roc::hipblas)
endif()
if (GGML_SYCL)
find_package(DNNL)
if (${DNNL_FOUND} AND GGML_SYCL_TARGET STREQUAL "INTEL")
list(APPEND GGML_SYCL_INTERFACE_LINK_LIBRARIES DNNL::dnnl)
endif()
if (WIN32)
find_package(IntelSYCL REQUIRED)
find_package(MKL REQUIRED)
list(APPEND GGML_SYCL_INTERFACE_LINK_LIBRARIES IntelSYCL::SYCL_CXX MKL::MKL MKL::MKL_SYCL)
endif()
endif()
endif()
set(_ggml_all_targets "")
foreach(_ggml_backend ${GGML_AVAILABLE_BACKENDS})
string(REPLACE "-" "_" _ggml_backend_pfx "${_ggml_backend}")
string(TOUPPER "${_ggml_backend_pfx}" _ggml_backend_pfx)
find_library(${_ggml_backend_pfx}_LIBRARY ${_ggml_backend}
REQUIRED
HINTS ${GGML_LIB_DIR}
NO_CMAKE_FIND_ROOT_PATH)
message(STATUS "Found ${${_ggml_backend_pfx}_LIBRARY}")
add_library(ggml::${_ggml_backend} UNKNOWN IMPORTED)
set_target_properties(ggml::${_ggml_backend}
PROPERTIES
INTERFACE_INCLUDE_DIRECTORIES "${GGML_INCLUDE_DIR}"
IMPORTED_LINK_INTERFACE_LANGUAGES "CXX"
IMPORTED_LOCATION "${${_ggml_backend_pfx}_LIBRARY}"
INTERFACE_COMPILE_FEATURES c_std_90
POSITION_INDEPENDENT_CODE ON)
string(REGEX MATCH "^ggml-cpu" is_cpu_variant "${_ggml_backend}")
if(is_cpu_variant)
list(APPEND GGML_CPU_INTERFACE_LINK_LIBRARIES "ggml::ggml" "ggml::ggml-base")
set_target_properties(ggml::${_ggml_backend}
PROPERTIES
INTERFACE_LINK_LIBRARIES "${GGML_CPU_INTERFACE_LINK_LIBRARIES}")
if(GGML_CPU_INTERFACE_LINK_OPTIONS)
set_target_properties(ggml::${_ggml_backend}
PROPERTIES
INTERFACE_LINK_OPTIONS "${GGML_CPU_INTERFACE_LINK_OPTIONS}")
endif()
else()
list(APPEND ${_ggml_backend_pfx}_INTERFACE_LINK_LIBRARIES "ggml::ggml" "ggml::ggml-base")
set_target_properties(ggml::${_ggml_backend}
PROPERTIES
INTERFACE_LINK_LIBRARIES "${${_ggml_backend_pfx}_INTERFACE_LINK_LIBRARIES}")
if(${_ggml_backend_pfx}_INTERFACE_LINK_OPTIONS)
set_target_properties(ggml::${_ggml_backend}
PROPERTIES
INTERFACE_LINK_OPTIONS "${${_ggml_backend_pfx}_INTERFACE_LINK_OPTIONS}")
endif()
endif()
list(APPEND _ggml_all_targets ggml::${_ggml_backend})
endforeach()
add_library(ggml::all INTERFACE IMPORTED)
set_target_properties(ggml::all
PROPERTIES
INTERFACE_LINK_LIBRARIES "${_ggml_all_targets}")
check_required_components(ggml)

2
ggml/include/ggml.h
View file

@ -1788,7 +1788,7 @@ extern "C" {
struct ggml_tensor * a,
int k);
#define GGML_KQ_MASK_PAD 32
#define GGML_KQ_MASK_PAD 64
// q: [n_embd, n_batch, n_head, 1]
// k: [n_embd, n_kv, n_head_kv, 1]

14
ggml/src/ggml-alloc.c
View file

@ -989,19 +989,7 @@ ggml_backend_buffer_t ggml_backend_alloc_ctx_tensors_from_buft(struct ggml_conte
this_size = GGML_PAD(ggml_backend_buft_get_alloc_size(buft, t), alignment);
}
if (this_size > max_size) {
GGML_LOG_ERROR("%s: tensor %s is too large to fit in a %s buffer (tensor size: %zu, max buffer size: %zu)\n",
__func__, t->name,
ggml_backend_buft_name(buft),
this_size, max_size);
for (size_t i = 0; i < n_buffers; i++) {
ggml_backend_buffer_free(buffers[i]);
}
free(buffers);
return NULL;
}
if ((cur_buf_size + this_size) > max_size) {
if (cur_buf_size > 0 && (cur_buf_size + this_size) > max_size) {
// allocate tensors in the current buffer
if (!alloc_tensor_range(ctx, first, t, buft, cur_buf_size, &buffers, &n_buffers)) {
return NULL;

169
ggml/src/ggml-cpu/ggml-cpu-quants.c
View file

@ -298,6 +298,90 @@ static const uint64_t table_b2b_0[1 << 8] = { B8(00, 10) }; // ( b) << 4
static const uint64_t table_b2b_1[1 << 8] = { B8(10, 00) }; // (!b) << 4
#endif
#if defined(__loongarch_sx)
static __m128i lsx_packs_w(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_w(a, 15);
tmp1 = __lsx_vsat_w(b, 15);
return __lsx_vpickev_h(tmp1, tmp);
}
static __m128i lsx_packs_h(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_h(a, 7);
tmp1 = __lsx_vsat_h(b, 7);
return __lsx_vpickev_b(tmp1, tmp);
}
static __m128i lsx_packus_h(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_hu(a, 7);
tmp1 = __lsx_vsat_hu(b, 7);
return __lsx_vpickev_b(tmp1, tmp);
}
static __m128i lsx_maddubs_h(__m128i a, __m128i b) {
__m128i tmp1, tmp2;
tmp1 = __lsx_vmulwev_h_b(a, b);
tmp2 = __lsx_vmulwod_h_b(a, b);
return __lsx_vsadd_h(tmp1, tmp2);
}
static __m128i lsx_madd_h(__m128i a, __m128i b) {
__m128i tmp1, tmp2;
tmp1 = __lsx_vmulwev_w_h(a, b);
tmp2 = __lsx_vmulwod_w_h(a, b);
return __lsx_vadd_w(tmp1, tmp2);
}
static __m128i lsx_set_w(int32_t a, int32_t b, int32_t c, int32_t d) {
v4i32 __ret = {d, c, b, a};
return (__m128i)__ret;
}
static __m128i lsx_shuffle_b(__m128i a, __m128i b) {
__m128i mask_f, zero, tmp0, tmp2, mask;
int f = 0x8f;
mask_f = __lsx_vreplgr2vr_b(f);
zero = __lsx_vldi(0);
tmp0 = __lsx_vand_v(b, mask_f); // get mask with low 4 bit and sign bits
tmp0 = __lsx_vori_b(tmp0, 0x10); // make each mask or with 0x10 prepare for positive
mask = __lsx_vsle_b(zero, tmp0); // if mask >= 0, set mask
tmp2 = __lsx_vand_v(tmp0, mask); // maskout the in2 < ones
return __lsx_vshuf_b(a, zero, tmp2);
}
static __m128i lsx_hadd_h(__m128i a, __m128i b) {
__m128i tmp1 = __lsx_vpickev_h(b, a);
__m128i tmp2 = __lsx_vpickod_h(b, a);
return __lsx_vadd_h(tmp1, tmp2);
}
static __m128i lsx_hadd_w(__m128i a, __m128i b) {
__m128i tmp1 = __lsx_vpickev_w(b, a);
__m128i tmp2 = __lsx_vpickod_w(b, a);
return __lsx_vadd_w(tmp1, tmp2);
}
static __m128 lsx_hadd_s(__m128 a, __m128 b) {
__m128 tmp1 = (__m128)__lsx_vpickev_w((__m128i)b, (__m128i)a);
__m128 tmp2 = (__m128)__lsx_vpickod_w((__m128i)b, (__m128i)a);
return __lsx_vfadd_s(tmp1, tmp2);
}
static inline float hsum_float_4x4(const __m128 a, const __m128 b, const __m128 c, const __m128 d) {
__m128 res_0 =lsx_hadd_s(a, b);
__m128 res_1 =lsx_hadd_s(c, d);
__m128 res =lsx_hadd_s(res_0, res_1);
res =lsx_hadd_s(res, res);
res =lsx_hadd_s(res, res);
return ((v4f32)res)[0];
}
#endif
#if defined(__loongarch_asx)
#ifdef __clang__
@ -396,11 +480,6 @@ static __m256i lasx_set_w(int e7, int e6, int e5, int e4, int e3, int e2, int e1
return (__m256i)__ret;
}
static __m128i lsx_set_w(int32_t a, int32_t b, int32_t c, int32_t d) {
v4i32 __ret = {d, c, b, a};
return (__m128i)__ret;
}
static __m256i lasx_set_d(int64_t a, int64_t b, int64_t c, int64_t d) {
v4i64 __ret = {d, c, b, a};
return (__m256i)__ret;
@ -410,18 +489,6 @@ static __m256i lasx_insertf128( __m128i x, __m128i y) {
return lasx_set_q(x, y);
}
static __m128i lsx_shuffle_b(__m128i a, __m128i b) {
__m128i mask_f, zero, tmp0, tmp2, mask;
int f = 0x8f;
mask_f = __lsx_vreplgr2vr_b(f);
zero = __lsx_vldi(0);
tmp0 = __lsx_vand_v(b, mask_f); // get mask with low 4 bit and sign bits
tmp0 = __lsx_vori_b(tmp0, 0x10); // make each mask or with 0x10 prepare for positive
mask = __lsx_vsle_b(zero, tmp0); // if mask >= 0, set mask
tmp2 = __lsx_vand_v(tmp0, mask); // maskout the in2 < ones
return __lsx_vshuf_b(a, zero, tmp2);
}
static __m256i lasx_shuffle_b(__m256i a, __m256i b) {
__m256i mask_f, zero, tmp0, tmp2, mask;
int f = 0x8f;
@ -483,25 +550,6 @@ static __m128 lasx_extractf128( __m256 a, int pos) {
return ret;
}
static __m128i lsx_hadd_h(__m128i a, __m128i b) {
__m128i tmp1 = __lsx_vpickev_h(b, a);
__m128i tmp2 = __lsx_vpickod_h(b, a);
return __lsx_vadd_h(tmp1, tmp2);
}
static __m128i lsx_hadd_w(__m128i a, __m128i b) {
__m128i tmp1 = __lsx_vpickev_w(b, a);
__m128i tmp2 = __lsx_vpickod_w(b, a);
return __lsx_vadd_w(tmp1, tmp2);
}
static __m128 lsx_hadd_s(__m128 a, __m128 b) {
__m128 tmp1 = (__m128)__lsx_vpickev_w((__m128i)b, (__m128i)a);
__m128 tmp2 = (__m128)__lsx_vpickod_w((__m128i)b, (__m128i)a);
return __lsx_vfadd_s(tmp1, tmp2);
}
static __m256i lasx_maddubs_h(__m256i a, __m256i b) {
__m256i tmp1, tmp2;
tmp1 = __lasx_xvmulwev_h_b(a, b);
@ -530,42 +578,6 @@ static __m256i lasx_packs_h(__m256i a, __m256i b) {
return __lasx_xvpickev_b(tmp1, tmp);
}
static __m128i lsx_packs_w(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_w(a, 15);
tmp1 = __lsx_vsat_w(b, 15);
return __lsx_vpickev_h(tmp1, tmp);
}
static __m128i lsx_packs_h(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_h(a, 7);
tmp1 = __lsx_vsat_h(b, 7);
return __lsx_vpickev_b(tmp1, tmp);
}
static __m128i lsx_packus_h(__m128i a, __m128i b) {
__m128i tmp, tmp1;
tmp = __lsx_vsat_hu(a, 7);
tmp1 = __lsx_vsat_hu(b, 7);
return __lsx_vpickev_b(tmp1, tmp);
}
static __m128i lsx_maddubs_h(__m128i a, __m128i b) {
__m128i tmp1, tmp2;
tmp1 = __lsx_vmulwev_h_b(a, b);
tmp2 = __lsx_vmulwod_h_b(a, b);
return __lsx_vsadd_h(tmp1, tmp2);
}
static __m128i lsx_madd_h(__m128i a, __m128i b) {
__m128i tmp1, tmp2;
tmp1 = __lsx_vmulwev_w_h(a, b);
tmp2 = __lsx_vmulwod_w_h(a, b);
return __lsx_vadd_w(tmp1, tmp2);
}
// multiply int8_t, add results pairwise twice
static inline __m128i mul_sum_i8_pairs(const __m128i x, const __m128i y) {
// Get absolute values of x vectors
@ -2233,21 +2245,22 @@ void ggml_vec_dot_q4_0_q8_0(int n, float * restrict s, size_t bs, const void * r
}
sumf = hsum_float_8(acc);
#elif defined(__loongarch_sx)
// set constants
const __m128i low_mask = __lsx_vreplgr2vr_b(0xF);
const __m128i off = __lsx_vreplgr2vr_b(8);
// Initialize accumulator with zeros
__m128 acc_0 = __lsx_vldi(0);
__m128 acc_1 = __lsx_vldi(0);
__m128 acc_2 = __lsx_vldi(0);
__m128 acc_3 = __lsx_vldi(0);
__m128 acc_0 = (__m128)__lsx_vldi(0);
__m128 acc_1 = (__m128)__lsx_vldi(0);
__m128 acc_2 = (__m128)__lsx_vldi(0);
__m128 acc_3 = (__m128)__lsx_vldi(0);
for (; ib + 1 < nb; ib += 2) {
// Compute combined scale for the block 0 and 1
const __m128 d_0_1 = __lsx_vreplgr2vr_w( GGML_FP16_TO_FP32(x[ib].d) * GGML_FP16_TO_FP32(y[ib].d) );
const __m128 d_0_1 = (__m128)__lsx_vreplgr2vr_w( GGML_FP16_TO_FP32(x[ib].d) * GGML_FP16_TO_FP32(y[ib].d) );
const __m128i tmp_0_1 = __lsx_vld((const __m128i *)x[ib].qs, 0);
@ -2265,7 +2278,7 @@ void ggml_vec_dot_q4_0_q8_0(int n, float * restrict s, size_t bs, const void * r
//_mm_prefetch(&y[ib] + 2 * sizeof(block_q8_0), _MM_HINT_T0);
// Compute combined scale for the block 2 and 3
const __m128 d_2_3 = __lsx_vreplgr2vr_w( GGML_FP16_TO_FP32(x[ib + 1].d) * GGML_FP16_TO_FP32(y[ib + 1].d) );
const __m128 d_2_3 = (__m128)__lsx_vreplgr2vr_w( GGML_FP16_TO_FP32(x[ib + 1].d) * GGML_FP16_TO_FP32(y[ib + 1].d) );
const __m128i tmp_2_3 = __lsx_vld((const __m128i *)x[ib + 1].qs, 0);

21
ggml/src/ggml-cuda/common.cuh
View file

@ -61,6 +61,13 @@
#define GGML_CUDA_CC_RDNA2 (GGML_CUDA_CC_OFFSET_AMD + 0x1030) // RX 6000, minimum for dp4a
#define GGML_CUDA_CC_RDNA3 (GGML_CUDA_CC_OFFSET_AMD + 0x1100) // RX 7000, minimum for WMMA
#define GGML_CUDA_CC_IS_RDNA(cc) (cc >= GGML_CUDA_CC_RDNA1)
#define GGML_CUDA_CC_IS_RDNA1(cc) (cc >= GGML_CUDA_CC_RDNA1 && cc < GGML_CUDA_CC_RDNA2)
#define GGML_CUDA_CC_IS_RDNA2(cc) (cc >= GGML_CUDA_CC_RDNA2 && cc < GGML_CUDA_CC_RDNA3)
#define GGML_CUDA_CC_IS_RDNA3(cc) (cc >= GGML_CUDA_CC_RDNA3)
#define GGML_CUDA_CC_IS_GCN(cc) (cc > GGML_CUDA_CC_OFFSET_AMD && cc < GGML_CUDA_CC_CDNA)
#define GGML_CUDA_CC_IS_CDNA(cc) (cc >= GGML_CUDA_CC_CDNA && cc < GGML_CUDA_CC_RDNA1)
#define GGML_CUDA_CC_QY1 210
#define GGML_CUDA_CC_QY2 220
@ -148,7 +155,7 @@ typedef float2 dfloat2;
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= GGML_CUDA_CC_VOLTA
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= GGML_CUDA_CC_TURING
#define INT8_MMA_AVAILABLE
#define NEW_MMA_AVAILABLE
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)) && __CUDA_ARCH__ >= GGML_CUDA_CC_TURING
#if !(defined(GGML_USE_MUSA) && __MUSA_ARCH__ <= GGML_CUDA_CC_QY1)
@ -159,14 +166,24 @@ static constexpr bool fast_fp16_available(const int cc) {
return cc >= GGML_CUDA_CC_PASCAL && cc != 610;
}
// Any FP16 tensor cores are available.
static constexpr bool fp16_mma_available(const int cc) {
return cc < GGML_CUDA_CC_OFFSET_AMD && cc >= GGML_CUDA_CC_VOLTA;
}
static constexpr bool int8_mma_available(const int cc) {
// Volta technically had FP16 tensor cores but they work very differently compared to Turing and later.
static constexpr bool new_mma_available(const int cc) {
return cc < GGML_CUDA_CC_OFFSET_AMD && cc >= GGML_CUDA_CC_TURING;
}
static constexpr __device__ int ggml_cuda_get_physical_warp_size() {
#if defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)
return __AMDGCN_WAVEFRONT_SIZE;
#else
return 32;
#endif // defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__)
}
[[noreturn]]
static __device__ void no_device_code(
const char * file_name, const int line, const char * function_name, const int arch, const char * arch_list) {

175
ggml/src/ggml-cuda/fattn-common.cuh
View file

@ -516,6 +516,114 @@ constexpr __device__ dequantize_1_f32_t get_dequantize_1_f32(ggml_type type_V) {
nullptr;
}
// The HIP compiler for some reason complains that it can't unroll a loop because of the jt*ncols + j >= ne01 conditional.
#ifdef __clang__
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wpass-failed"
#endif // __clang__
template<int D, int ncols, int KQ_stride> // D == head size
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
__launch_bounds__(D, 1)
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
static __global__ void flash_attn_stream_k_fixup(
float * __restrict__ dst, const float2 * __restrict__ dst_fixup, const int ne01, const int ne02, const int ne11) {
const float * dst_fixup_data = ((const float *) dst_fixup) + gridDim.x*(2*2*ncols);
const int iter_k = ne11 / KQ_stride;
const int iter_j = (ne01 + (ncols - 1)) / ncols;
const int bidx0 = blockIdx.x;
const int kbc0 = (bidx0 + 0)*iter_k*iter_j*ne02 / gridDim.x;
const int kbc0_stop = (bidx0 + 1)*iter_k*iter_j*ne02 / gridDim.x;
const bool did_not_have_any_data = kbc0 == kbc0_stop;
const bool wrote_beginning_of_tile = kbc0 % iter_k == 0;
const bool did_not_write_last = kbc0/iter_k == kbc0_stop/iter_k && kbc0_stop % iter_k != 0;
if (did_not_have_any_data || wrote_beginning_of_tile || did_not_write_last) {
return;
}
const int channel = kbc0 / (iter_k*iter_j);
const int jt = (kbc0 - channel*iter_k*iter_j) / iter_k;
dst += jt*ncols*ne02*D + channel*D;
// Load the partial result that needs a fixup:
float dst_val[ncols] = {0.0f};
float max_val[ncols] = {0.0f};
float rowsum[ncols] = {0.0f};
#pragma unroll
for (int j = 0; j < ncols; ++j) {
if (jt*ncols + j >= ne01) {
break;
}
dst_val[j] = dst[j*ne02*D + threadIdx.x];
const float2 tmp = dst_fixup[bidx0*ncols + j];
max_val[j] = tmp.x;
rowsum[j] = tmp.y;
}
// Iterate over previous blocks and compute the combined results.
// All CUDA blocks that get here must have a previous block that needs a fixup.
int bidx = bidx0 - 1;
int kbc_stop = kbc0;
while(true) {
const int kbc = bidx*iter_k*iter_j*ne02 / gridDim.x;
if (kbc == kbc_stop) { // Did not have any data.
bidx--;
kbc_stop = kbc;
continue;
}
#pragma unroll
for (int j = 0; j < ncols; ++j) {
if (jt*ncols + j >= ne01) {
break;
}
const float dst_add = dst_fixup_data[bidx*ncols*D + j*D + threadIdx.x];
const float2 tmp = dst_fixup[(gridDim.x + bidx)*ncols + j];
// Scale the current and new value accumulators depending on the max. values.
const float max_val_new = fmaxf(max_val[j], tmp.x);
const float diff_val = max_val[j] - max_val_new;
const float diff_add = tmp.x - max_val_new;
const float scale_val = diff_val >= SOFTMAX_FTZ_THRESHOLD ? expf(diff_val) : 0.0f;
const float scale_add = diff_add >= SOFTMAX_FTZ_THRESHOLD ? expf(diff_add) : 0.0f;
dst_val[j] = scale_val*dst_val[j] + scale_add*dst_add;
rowsum[j] = scale_val*rowsum[j] + scale_add*tmp.y;
max_val[j] = max_val_new;
}
// If this block started in a previous tile we are done and don't need to combine additional partial results.
if (kbc % iter_k == 0 || kbc/iter_k < kbc0/iter_k) {
break;
}
bidx--;
kbc_stop = kbc;
}
// Write back final result:
#pragma unroll
for (int j = 0; j < ncols; ++j) {
if (jt*ncols + j >= ne01) {
return;
}
dst[j*ne02*D + threadIdx.x] = dst_val[j] / rowsum[j];
}
}
#ifdef __clang__
#pragma clang diagnostic pop
#endif // __clang__
template<int D, int parallel_blocks> // D == head size
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
__launch_bounds__(D, 1)
@ -581,10 +689,11 @@ static void on_no_fattn_vec_case(const int D) {
}
}
template <int D, int parallel_blocks>
// parallel_blocks == 0 is stream-k decomposition
template <int D, int cols_per_block, int parallel_blocks, int KQ_stride>
void launch_fattn(
ggml_backend_cuda_context & ctx, ggml_tensor * dst, fattn_kernel_t fattn_kernel,
const int nwarps, const int cols_per_block, const bool need_f16_K, const bool need_f16_V
const int nwarps, const size_t nbytes_shared, const bool need_f16_K, const bool need_f16_V
) {
const ggml_tensor * Q = dst->src[0];
const ggml_tensor * K = dst->src[1];
@ -603,20 +712,23 @@ void launch_fattn(
GGML_ASSERT(K->ne[1] % FATTN_KQ_STRIDE == 0 && "Incorrect KV cache padding.");
GGML_ASSERT(Q->ne[3] == 1);
ggml_cuda_pool & pool = ctx.pool();
cudaStream_t main_stream = ctx.stream();
const int nsm = ggml_cuda_info().devices[ggml_cuda_get_device()].nsm;
ggml_cuda_pool_alloc<half> K_f16(pool);
ggml_cuda_pool_alloc<half> V_f16(pool);
ggml_cuda_pool_alloc<float> dst_tmp(pool);
ggml_cuda_pool_alloc<float2> dst_tmp_meta(pool);
char * K_data = (char *) K->data;
const char * K_data = (const char *) K->data;
size_t nb11 = K->nb[1];
size_t nb12 = K->nb[2];
size_t nb13 = K->nb[3];
char * V_data = (char *) V->data;
const char * V_data = (const char *) V->data;
size_t nb21 = V->nb[1];
size_t nb22 = V->nb[2];
size_t nb23 = V->nb[3];
@ -649,39 +761,60 @@ void launch_fattn(
nb23 = nb23*bs*sizeof(half)/ts;
}
const int ntiles_x = ((Q->ne[1] + cols_per_block - 1) / cols_per_block);
const int ntiles_total = ntiles_x*Q->ne[2]*Q->ne[3];
const dim3 block_dim(WARP_SIZE, nwarps, 1);
dim3 blocks_num;
if (parallel_blocks == 0) {
// For short contexts it can be faster to have the SMs work on whole tiles because this lets us skip the fixup.
const int tiles_nwaves = (ntiles_total - nsm - 1) / nsm;
const bool tiles_inefficient = 3*nsm < 2*tiles_nwaves*ntiles_total;
const bool short_context = K->ne[1] < 4096;
const int nblocks_stream_k = 2*nsm;
blocks_num.x = short_context && !tiles_inefficient ? ntiles_total : nblocks_stream_k;
blocks_num.y = 1;
blocks_num.z = 1;
dst_tmp_meta.alloc(blocks_num.x*cols_per_block * (2*2 + D) * sizeof(float));
} else {
blocks_num.x = parallel_blocks*ntiles_x;
blocks_num.y = Q->ne[2];
blocks_num.z = Q->ne[3];
if (parallel_blocks > 1) {
dst_tmp.alloc(parallel_blocks*ggml_nelements(KQV));
dst_tmp_meta.alloc(parallel_blocks*ggml_nrows(KQV));
}
}
const dim3 block_dim(WARP_SIZE, nwarps, 1);
const dim3 blocks_num(parallel_blocks*((Q->ne[1] + cols_per_block - 1) / cols_per_block), Q->ne[2], Q->ne[3]);
const int shmem = 0;
float scale = 1.0f;
float max_bias = 0.0f;
float logit_softcap = 0.0f;
memcpy(&scale, (float *) KQV->op_params + 0, sizeof(float));
memcpy(&max_bias, (float *) KQV->op_params + 1, sizeof(float));
memcpy(&logit_softcap, (float *) KQV->op_params + 2, sizeof(float));
memcpy(&scale, (const float *) KQV->op_params + 0, sizeof(float));
memcpy(&max_bias, (const float *) KQV->op_params + 1, sizeof(float));
memcpy(&logit_softcap, (const float *) KQV->op_params + 2, sizeof(float));
if (logit_softcap != 0.0f) {
scale /= logit_softcap;
}
const uint32_t n_head = Q->ne[2];
const uint32_t n_head_log2 = 1u << (uint32_t) floorf(log2f((float) n_head));
const uint32_t n_head_log2 = 1u << uint32_t(floorf(log2f(float(n_head))));
const float m0 = powf(2.0f, -(max_bias ) / n_head_log2);
const float m1 = powf(2.0f, -(max_bias / 2.0f) / n_head_log2);
fattn_kernel<<<blocks_num, block_dim, shmem, main_stream>>>(
fattn_kernel<<<blocks_num, block_dim, nbytes_shared, main_stream>>>(
(const char *) Q->data,
K_data,
V_data,
mask ? ((const char *) mask->data) : nullptr,
(parallel_blocks) == 1 ? (float *) KQV->data : dst_tmp.ptr, dst_tmp_meta.ptr,
(parallel_blocks) > 1 ? dst_tmp.ptr : (float *) KQV->data, dst_tmp_meta.ptr,
scale, max_bias, m0, m1, n_head_log2, logit_softcap,
Q->ne[0], Q->ne[1], Q->ne[2], Q->ne[3],
K->ne[0], K->ne[1], K->ne[2], K->ne[3],
@ -693,16 +826,22 @@ void launch_fattn(
);
CUDA_CHECK(cudaGetLastError());
if ((parallel_blocks) == 1) {
return;
}
if constexpr (parallel_blocks == 0) {
if (blocks_num.x % ntiles_total != 0) { // Fixup is only needed if the SMs work on fractional tiles.
const dim3 block_dim_combine(D, 1, 1);
const dim3 blocks_num_combine = blocks_num;
flash_attn_stream_k_fixup<D, cols_per_block, KQ_stride>
<<<blocks_num_combine, block_dim_combine, 0, main_stream>>>
((float *) KQV->data, dst_tmp_meta.ptr, Q->ne[1], Q->ne[2], K->ne[1]);
}
} else if constexpr (parallel_blocks > 1) {
const dim3 block_dim_combine(D, 1, 1);
const dim3 blocks_num_combine(Q->ne[1], blocks_num.y, blocks_num.z);
const int shmem_combine = 0;
flash_attn_combine_results<D, parallel_blocks>
<<<blocks_num_combine, block_dim_combine, shmem_combine, main_stream>>>
<<<blocks_num_combine, block_dim_combine, 0, main_stream>>>
(dst_tmp.ptr, dst_tmp_meta.ptr, (float *) KQV->data);
}
CUDA_CHECK(cudaGetLastError());
}

637
ggml/src/ggml-cuda/fattn-mma-f16.cuh Normal file
View file

@ -0,0 +1,637 @@
#include "common.cuh"
#include "mma.cuh"
#include "fattn-common.cuh"
template<int D, int ncols, int nwarps, int KQ_stride, bool use_logit_softcap, bool needs_fixup, bool is_fixup>
static __device__ __forceinline__ void flash_attn_ext_f16_process_tile(
const float2 * const __restrict__ Q_f2,
const half2 * const __restrict__ K_h2,
const half2 * const __restrict__ V_h2,
const half * const __restrict__ maskh,
float2 * const __restrict__ dstk,
float2 * const __restrict__ dstk_fixup,
const float scale,
const float slope,
const float logit_softcap,
const int ne00,
const int ne01,
const int ne02,
const int ne03,
const int ne10,
const int ne11,
const int ne12,
const int ne13,
const int ne31,
const int nb31,
const int nb01,
const int nb02,
const int nb03,
const int nb11,
const int nb12,
const int nb13,
const int nb21,
const int nb22,
const int nb23,
const int ne0,
const int ne1,
const int ne2,
const int ne3,
const int jt,
const int kb0_start,
const int kb0_stop) {
#ifdef NEW_MMA_AVAILABLE
//In this kernel Q, K, V are matrices while i, j, k are matrix indices.
typedef mma_A_I16K8<half2> mma_A;
typedef mma_B_J8K8<half2> mma_B;
typedef mma_C_I16J8<float> mma_C_KQ;
typedef mma_C_I16J8<half2> mma_C_VKQ;
static_assert(nwarps*mma_B::J % ncols == 0, "bad nwarps");
constexpr int np = nwarps*mma_B::J / ncols; // Number of parallel CUDA warps per Q column.
static_assert(D % nwarps == 0, "bad D");
static_assert(KQ_stride % nwarps == 0, "bad KQ_stride");
constexpr int D2_padded = D/2 + 4; // Size of D in half2, padded to avoid shared memory bank conflicts.
extern __shared__ half2 tile_KV[]; // Temporary shared buffer for loading K/V data with KQ_stride*D logical elements.
const int stride_Q = nb01 / sizeof(float2);
const int stride_KV = nb11 / sizeof(half2);
const int stride_mask = nb31 / sizeof(half);
mma_B Q_B[D/(2*mma_B::K)];
mma_C_VKQ VKQ_C[D/mma_C_VKQ::I];
float2 KQ_rowsum = {0.0f, 0.0f};
float2 KQ_max = {-FLT_MAX/2.0f, -FLT_MAX/2.0f};
float2 KQ_max_scale = {0.0f, 0.0f};
// Temporarily load Q data into tile_KV, will be loaded into registers afterwards.
// The loading is done with decreasing granularity for D for better memory bandwidth.
const half2 scale_h2 = make_half2(scale, scale);
#pragma unroll
for (int stride_k : {WARP_SIZE, WARP_SIZE/2, WARP_SIZE/4}) {
const int k0_start = stride_k == WARP_SIZE ? 0 : D/2 - (D/2) % (2*stride_k);
const int k0_stop = D/2 - (D/2) % (1*stride_k);
const int stride_j = WARP_SIZE / stride_k;
if (nwarps*stride_j > ncols && threadIdx.y*stride_j >= ncols) {
break;
}
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps*stride_j) {
const int j = j0 + threadIdx.y*stride_j + (stride_k == WARP_SIZE ? 0 : threadIdx.x / stride_k);
if (jt*ncols + j < ne01) {
#pragma unroll
for (int k0 = k0_start; k0 < k0_stop; k0 += stride_k) {
const int k = k0 + (stride_k == WARP_SIZE ? threadIdx.x : threadIdx.x % stride_k);
const float2 tmp = Q_f2[(jt*ncols + j)*stride_Q + k];
tile_KV[j*D2_padded + k] = scale_h2 * make_half2(tmp.x, tmp.y);
}
} else {
#pragma unroll
for (int k0 = k0_start; k0 < k0_stop; k0 += stride_k) {
const int k = k0 + (stride_k == WARP_SIZE ? threadIdx.x : threadIdx.x % stride_k);
tile_KV[j*D2_padded + k] = make_half2(0.0f, 0.0f);
}
}
}
}
__syncthreads();
{
const int j0 = (threadIdx.y / np) * mma_B::J;
#pragma unroll
for (int k0 = 0; k0 < D/2; k0 += mma_B::K) {
Q_B[k0/mma_B::K].load_ldmatrix(tile_KV + j0*D2_padded + k0, D2_padded);
}
}
__syncthreads();
// Iterate over ne11 == previous tokens:
for (int kb0 = kb0_start; kb0 < kb0_stop; ++kb0) {
const int k_VKQ_0 = kb0*KQ_stride;
mma_C_KQ KQ_C[KQ_stride/(np*mma_C_KQ::I)];
// Load K data into tile with decreasing granularity for D for better memory bandwidth:
static_assert(KQ_stride % (4*nwarps) == 0, "out of bounds");
#pragma unroll
for (int stride_k : {WARP_SIZE, WARP_SIZE/2, WARP_SIZE/4}) {
const int k0_start = stride_k == WARP_SIZE ? 0 : D/2 - (D/2) % (2*stride_k);
const int k0_stop = D/2 - (D/2) % (1*stride_k);
const int stride_i = WARP_SIZE / stride_k;
#pragma unroll
for (int i_KQ_0 = 0; i_KQ_0 < KQ_stride; i_KQ_0 += nwarps*stride_i) {
const int i_KQ = i_KQ_0 + threadIdx.y*stride_i + (stride_k == WARP_SIZE ? 0 : threadIdx.x / stride_k);
#pragma unroll
for (int k_KQ_0 = k0_start; k_KQ_0 < k0_stop; k_KQ_0 += stride_k) {
const int k_KQ = k_KQ_0 + (stride_k == WARP_SIZE ? threadIdx.x : threadIdx.x % stride_k);
tile_KV[i_KQ*D2_padded + k_KQ] = K_h2[(k_VKQ_0 + i_KQ)*stride_KV + k_KQ];
}
}
}
__syncthreads();
// Calculate tile of KQ:
#pragma unroll
for (int i_KQ_00 = 0; i_KQ_00 < KQ_stride; i_KQ_00 += np*mma_A::I) {
const int i_KQ_0 = i_KQ_00 + (threadIdx.y % np)*mma_A::I;
#pragma unroll
for (int k_KQ_0 = 0; k_KQ_0 < D/2; k_KQ_0 += mma_A::K) {
mma_A K_A;
K_A.load_ldmatrix(tile_KV + i_KQ_0*D2_padded + k_KQ_0, D2_padded);
KQ_C[i_KQ_00/(np*mma_A::I)].mma(K_A, Q_B[k_KQ_0/mma_A::K]);
}
}
__syncthreads();
if (use_logit_softcap) {
static_assert(KQ_stride % (np*mma_C_KQ::I) == 0, "bad loop size");
#pragma unroll
for (int i = 0; i < KQ_stride/(np*mma_C_KQ::I); ++i) {
#pragma unroll
for (int l = 0; l < mma_C_KQ::ne; ++l) {
KQ_C[i].x[l] = logit_softcap*tanhf(KQ_C[i].x[l]);
}
}
}
if (maskh) {
static_assert(KQ_stride % (np *mma_C_KQ::I) == 0, "bad loop size");
static_assert(ncols % (nwarps/np*mma_C_KQ::J) == 0, "bad loop size");
#pragma unroll
for (int i00 = 0; i00 < KQ_stride; i00 += np*mma_C_KQ::I) {
const int i0 = i00 + (threadIdx.y % np)*mma_C_KQ::I;
#pragma unroll
for (int l = 0; l < mma_C_KQ::ne; ++l) {
const int i = i0 + mma_C_KQ::get_i(l);
const int j = (threadIdx.y / np)*mma_C_KQ::J + mma_C_KQ::get_j(l);
KQ_C[i00/(np*mma_C_KQ::I)].x[l] += slope*__half2float(maskh[j*stride_mask + k_VKQ_0 + i]);
}
}
}
// Calculate softmax for each KQ column using the current max. value.
// The divisor is stored in KQ_rowsum and will be applied at the end.
float2 KQ_max_new = KQ_max;
static_assert(KQ_stride % (np*mma_C_KQ::I) == 0, "bad loop size");
#pragma unroll
for (int k = 0; k < KQ_stride/(np*mma_C_KQ::I); ++k) {
#pragma unroll
for (int l0 = 0; l0 < mma_C_KQ::ne; l0 += 2) {
KQ_max_new.x = fmaxf(KQ_max_new.x, KQ_C[k].x[l0 + 0]);
KQ_max_new.y = fmaxf(KQ_max_new.y, KQ_C[k].x[l0 + 1]);
}
}
// Values per KQ column are spread across 8 threads, does not need full warp reduce:
#pragma unroll
for (int offset = 16; offset > 2; offset >>= 1) {
KQ_max_new.x = fmaxf(KQ_max_new.x, __shfl_xor_sync(0xFFFFFFFF, KQ_max_new.x, offset, WARP_SIZE));
KQ_max_new.y = fmaxf(KQ_max_new.y, __shfl_xor_sync(0xFFFFFFFF, KQ_max_new.y, offset, WARP_SIZE));
}
{
const float2 diff = make_float2(KQ_max.x - KQ_max_new.x, KQ_max.y - KQ_max_new.y);
KQ_max_scale = make_float2(expf(diff.x), expf(diff.y));
if (diff.x <= SOFTMAX_FTZ_THRESHOLD) {
KQ_max_scale.x = 0.0f;
}
if (diff.y <= SOFTMAX_FTZ_THRESHOLD) {
KQ_max_scale.y = 0.0f;
}
KQ_max = KQ_max_new;
}
float2 KQ_rowsum_add = make_float2(0.0f, 0.0f);
static_assert(KQ_stride % (np*mma_C_KQ::I) == 0, "bad loop size");
#pragma unroll
for (int k = 0; k < KQ_stride/(np*mma_C_KQ::I); ++k) {
#pragma unroll
for (int l = 0; l < mma_C_KQ::ne; ++l) {
const float KQ_max_l = l % 2 == 0 ? KQ_max.x : KQ_max.y;
const float diff = KQ_C[k].x[l] - KQ_max_l;
KQ_C[k].x[l] = expf(diff);
if (diff <= SOFTMAX_FTZ_THRESHOLD) {
KQ_C[k].x[l] = 0.0f;
}
if (l % 2 == 0) {
KQ_rowsum_add.x += KQ_C[k].x[l];
} else {
KQ_rowsum_add.y += KQ_C[k].x[l];
}
}
}
// Scale previous KQ_rowsum to account for a potential increase in KQ_max:
KQ_rowsum.x = KQ_max_scale.x*KQ_rowsum.x + KQ_rowsum_add.x;
KQ_rowsum.y = KQ_max_scale.y*KQ_rowsum.y + KQ_rowsum_add.y;
const half2 KQ_max_scale_h2 = make_half2(KQ_max_scale.x, KQ_max_scale.y);
#pragma unroll
for (int i = 0; i < D/mma_C_VKQ::I; ++i) {
#pragma unroll
for (int l = 0; l < mma_C_VKQ::ne; ++l) {
VKQ_C[i].x[l] *= KQ_max_scale_h2;
}
}
// Convert KQ C tiles into B tiles for VKQ calculation:
mma_B B[KQ_stride/(np*2*mma_B::K)];
static_assert(KQ_stride % (np*2*mma_B::K) == 0, "bad loop size");
#pragma unroll
for (int k = 0; k < KQ_stride/(np*2*mma_B::K); ++k) {
B[k] = KQ_C[k].to_mma_B();
}
// Load V data into tile with decreasing granularity for D for better memory bandwidth:
static_assert(KQ_stride % (4*nwarps) == 0, "out of bounds");
#pragma unroll
for (int stride_i : {WARP_SIZE, WARP_SIZE/2, WARP_SIZE/4}) {
const int i0_start = stride_i == WARP_SIZE ? 0 : D/2 - (D/2) % (2*stride_i);
const int i0_stop = D/2 - (D/2) % (1*stride_i);
const int stride_k = WARP_SIZE / stride_i;
#pragma unroll
for (int k_V_0 = 0; k_V_0 < KQ_stride; k_V_0 += nwarps*stride_k) {
const int k_V = k_V_0 + threadIdx.y*stride_k + (stride_i == WARP_SIZE ? 0 : threadIdx.x / stride_i);
#pragma unroll
for (int i_V_0 = i0_start; i_V_0 < i0_stop; i_V_0 += stride_i) {
const int i_V = i_V_0 + (stride_i == WARP_SIZE ? threadIdx.x : threadIdx.x % stride_i);
tile_KV[k_V*D2_padded + i_V] = V_h2[(k_VKQ_0 + k_V)*stride_KV + i_V];
}
}
}
__syncthreads();
// Calculate VKQ tile:
#pragma unroll
for (int i_VKQ_0 = 0; i_VKQ_0 < D; i_VKQ_0 += mma_C_VKQ::I) {
static_assert((KQ_stride/2) % (np*mma_A::K) == 0, "bad loop size");
#pragma unroll
for (int k00 = 0; k00 < KQ_stride/2; k00 += np*mma_A::K) {
const int k0 = k00 + (threadIdx.y % np)*mma_A::K;
mma_A A;
A.load_ldmatrix_trans(tile_KV + 2*k0*D2_padded + i_VKQ_0/2, D2_padded);
VKQ_C[i_VKQ_0/mma_C_VKQ::I].mma(A, B[k00/(np*mma_A::K)]);
}
}
__syncthreads();
}
// Finally, sum up partial KQ rowsums.
// The partial sums are spread across 8 threads each, does not need full reduce.
#pragma unroll
for (int offset = 16; offset > 2; offset >>= 1) {
KQ_rowsum.x += __shfl_xor_sync(0xFFFFFFFF, KQ_rowsum.x, offset, WARP_SIZE);
KQ_rowsum.y += __shfl_xor_sync(0xFFFFFFFF, KQ_rowsum.y, offset, WARP_SIZE);
}
// Write VKQ accumulators to shared memory in column-major format.
// It's faster to do small writes to shared memory, then large write to VRAM than to do small writes to VRAM.
// Also for np > 1 the combination is done via these values in shared memory.
const int j_cwd = threadIdx.y*mma_B::J + mma_B::get_j(-1); // j combine write data
#pragma unroll
for (int k0 = 0; k0 < D/2; k0 += mma_B::K) {
const mma_B B = VKQ_C[k0/mma_B::K].to_mma_B(); // Conversion of C to B matrix puts it in column-major format.
#pragma unroll
for (int l = 0; l < mma_B::ne; ++l) {
const int k = k0 + mma_B::get_k(l);
tile_KV[j_cwd*D2_padded + k] = B.x[l];
}
}
const int j_cwmo = (threadIdx.x % (2*mma_C_VKQ::J)) / mma_C_VKQ::J; // j combine write meta offset
const int j_cwm = threadIdx.y*(2*mma_C_VKQ::J) + 2*mma_C_VKQ::get_j(-1) + j_cwmo; // j combine write meta
const float2 KQ_cmr = make_float2(((const float *) &KQ_max)[j_cwmo], ((const float *) &KQ_rowsum)[j_cwmo]); // KQ combine max rowsum
if (((!needs_fixup && !is_fixup) || np > 1) && threadIdx.x < 2*mma_C_VKQ::J) {
// Use the 16 bytes of padding in each row to store the meta data: KQ max, KQ rowsum, KQ max scale.
((float2 *) tile_KV)[j_cwm*(D2_padded/2) + D/4] = KQ_cmr;
}
__syncthreads();
static_assert(np == 1 || np == 2 || np == 4, "bad np");
if (np == 1) {
// No combination is needed, the meta data can be directly written from registers to VRAM.
if (needs_fixup && threadIdx.x < mma_B::J) {
float2 * dstk_fixup_meta = dstk_fixup + blockIdx.x*ncols;
dstk_fixup_meta[j_cwm] = KQ_cmr;
}
if (is_fixup && threadIdx.x < mma_B::J) {
float2 * dstk_fixup_meta = dstk_fixup + (gridDim.x + blockIdx.x)*ncols;
dstk_fixup_meta[j_cwm] = KQ_cmr;
}
} else if (threadIdx.y % np == 0) {
// Combine the meta data for parallel warps via shared memory.
// Warps with threadIdx.y % np != 0 must NOT return early.
// All threads must return simultaneously to avoid race conditions with work on the next tile.
float * meta_j = (float *) tile_KV + (threadIdx.y*mma_B::J + threadIdx.x)*D2_padded + D/2;
float KQ_cm = -FLT_MAX/2; // KQ combine max per parallel warp.
if (np*mma_B::J == WARP_SIZE || threadIdx.x < np*mma_B::J) {
KQ_cm = meta_j[0];
}
float KQ_cmn = KQ_cm; // KQ combine max new, max between all parallel warps.
#pragma unroll
for (int offset = np*mma_B::J/2; offset >= mma_B::J; offset >>= 1) {
KQ_cmn = fmaxf(KQ_cmn, __shfl_xor_sync(0xFFFFFFFF, KQ_cmn, offset, WARP_SIZE));
}
const float KQ_cms = expf(KQ_cm - KQ_cmn); // KQ combine max scale per warp.
float KQ_crs = 0.0f; // KQ combine rowsum, scaled sum of all parallel warps.
if (np*mma_B::J == WARP_SIZE || threadIdx.x < np*mma_B::J) {
KQ_crs = KQ_cms*meta_j[1];
}
#pragma unroll
for (int offset = np*mma_B::J/2; offset >= mma_B::J; offset >>= 1) {
KQ_crs += __shfl_xor_sync(0xFFFFFFFF, KQ_crs, offset, WARP_SIZE);
}
// Write back combined meta data:
if (np*mma_B::J == WARP_SIZE || threadIdx.x < np*mma_B::J) {
meta_j[0] = KQ_cmn; // Combined max. KQ values.
meta_j[1] = KQ_crs; // Combined KQ rowsums.
meta_j[2] = KQ_cms; // KQ max scales per parallel warp.
}
if (needs_fixup && threadIdx.x < mma_B::J) {
float2 * dstk_fixup_meta = dstk_fixup + blockIdx.x*ncols;
dstk_fixup_meta[(threadIdx.y/np)*mma_B::J + threadIdx.x] = make_float2(KQ_cmn, KQ_crs);
}
if (is_fixup && threadIdx.x < mma_B::J) {
float2 * dstk_fixup_meta = dstk_fixup + (gridDim.x + blockIdx.x)*ncols;
dstk_fixup_meta[(threadIdx.y/np)*mma_B::J + threadIdx.x] = make_float2(KQ_cmn, KQ_crs);
}
}
if (np > 1) {
__syncthreads();
}
if (np == 1 || threadIdx.y % np == 0) {
// The first 2*2*gridDim.x*ncols floats in dstk_fixup are for storing max. values and row sums.
// The values after that are for the partial results of the individual blocks.
float2 * dstk_fixup_data = dstk_fixup + gridDim.x*(2*ncols) + blockIdx.x*(ncols*(D/2));
#pragma unroll
for (int stride_k : {WARP_SIZE, WARP_SIZE/2, WARP_SIZE/4}) {
const int k0_start = stride_k == WARP_SIZE ? 0 : D/2 - (D/2) % (2*stride_k);
const int k0_stop = D/2 - (D/2) % (1*stride_k);
const int stride_j = WARP_SIZE / stride_k;
if (nwarps*stride_j > ncols && threadIdx.y*stride_j >= ncols) {
break;
}
#pragma unroll
for (int j0_dst = 0; j0_dst < ncols; j0_dst += (nwarps/np)*stride_j) {
const int j_dst = j0_dst + (threadIdx.y/np)*stride_j + (stride_k == WARP_SIZE ? 0 : threadIdx.x / stride_k);
const int j_tile_KV = (j_dst/mma_B::J)*(np*mma_B::J) + j_dst % mma_B::J;
if (!is_fixup && jt*ncols + j_dst >= ne01) {
continue;
}
const float * meta_j = (const float *) tile_KV + j_tile_KV*D2_padded + D/2;
#pragma unroll
for (int k0 = k0_start; k0 < k0_stop; k0 += stride_k) {
const int k = k0 + (stride_k == WARP_SIZE ? threadIdx.x : threadIdx.x % stride_k);
float2 dstk_val = make_float2(0.0f, 0.0f);
#pragma unroll
for (int ip = 0; ip < np; ++ip) {
const float KQ_crs = np == 1 ? 1.0f : meta_j[ip*mma_B::J*D2_padded + 2];
const float2 dstk_val_add = __half22float2(tile_KV[(j_tile_KV + ip*mma_B::J)*D2_padded + k]);
dstk_val.x += dstk_val_add.x*KQ_crs;
dstk_val.y += dstk_val_add.y*KQ_crs;
}
if (!needs_fixup && !is_fixup) {
const float KQ_rowsum_j = meta_j[1];
dstk_val.x /= KQ_rowsum_j;
dstk_val.y /= KQ_rowsum_j;
}
if (is_fixup) {
dstk_fixup_data[j_dst*(D/2) + k] = dstk_val;
} else {
dstk[(jt*ncols + j_dst)*ne02*(D/2) + k] = dstk_val;
}
}
}
}
}
if (np > 1) {
__syncthreads();
}
#else
NO_DEVICE_CODE;
#endif // NEW_MMA_AVAILABLE
}
template<int D, int ncols, int nwarps, int KQ_stride, bool use_logit_softcap>
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
__launch_bounds__(nwarps*WARP_SIZE, 2)
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
static __global__ void flash_attn_ext_f16(
const char * __restrict__ Q,
const char * __restrict__ K,
const char * __restrict__ V,
const char * __restrict__ mask,
float * __restrict__ dst,
float2 * __restrict__ dst_meta,
const float scale,
const float max_bias,
const float m0,
const float m1,
const uint32_t n_head_log2,
const float logit_softcap,
const int ne00,
const int ne01,
const int ne02,
const int ne03,
const int ne10,
const int ne11,
const int ne12,
const int ne13,
const int ne31,
const int nb31,
const int nb01,
const int nb02,
const int nb03,
const int nb11,
const int nb12,
const int nb13,
const int nb21,
const int nb22,
const int nb23,
const int ne0,
const int ne1,
const int ne2,
const int ne3) {
// Skip unused kernel variants for faster compilation:
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
return;
}
static_assert(FATTN_KQ_STRIDE % KQ_stride == 0, "bad KQ_stride");
const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
const int iter_k = ne11 / KQ_stride;
const int iter_j = (ne01 + (ncols - 1)) / ncols;
// kbc == k block continuous, current index in continuous ijk space.
int kbc = (blockIdx.x + 0)*iter_k*iter_j*ne02 / gridDim.x;
const int kbc_stop = (blockIdx.x + 1)*iter_k*iter_j*ne02 / gridDim.x;
// If the seams of 2 CUDA blocks fall within an output tile their results need to be combined.
// For this we need to track both the block that starts the tile (needs_fixup) and the block that finishes the tile (is_fixup).
// In the most general case >2 seams can fall into the same tile.
// kb0 == k start index when in the output tile.
int kb0_start = kbc % iter_k;
int kb0_stop = min(iter_k, kb0_start + kbc_stop - kbc);
while (kbc < kbc_stop && kb0_stop == iter_k) {
const int channel = kbc / (iter_k*iter_j);
const int jt = (kbc - channel*iter_k*iter_j) / iter_k; // j index of current tile.
const float2 * Q_f2 = (const float2 *) (Q + nb02* channel);
const half2 * K_h2 = (const half2 *) (K + nb12*(channel / gqa_ratio));
const half2 * V_h2 = (const half2 *) (V + nb12*(channel / gqa_ratio)); // K and V have same shape
const half * maskh = mask ? (const half *) mask + (nb31/sizeof(half))*jt*ncols : nullptr;
float2 * dstk = ((float2 *) dst) + channel*(D/2);
const float slope = get_alibi_slope(max_bias, channel, n_head_log2, m0, m1);
constexpr bool is_fixup = false; // All but (potentially) the last iterations write their data to dst rather than the fixup buffer.
if (kb0_start == 0) {
constexpr bool needs_fixup = false; // CUDA block is working on an entire tile.
flash_attn_ext_f16_process_tile<D, ncols, nwarps, KQ_stride, use_logit_softcap, needs_fixup, is_fixup>
(Q_f2, K_h2, V_h2, maskh, dstk, dst_meta, scale, slope, logit_softcap,
ne00, ne01, ne02, ne03, ne10, ne11, ne12, ne13, ne31, nb31, nb01, nb02, nb03, nb11, nb12, nb13, nb21, nb22, nb23, ne0, ne1, ne2, ne3,
jt, kb0_start, kb0_stop);
} else {
constexpr bool needs_fixup = true; // CUDA block is working on the beginning of a tile.
flash_attn_ext_f16_process_tile<D, ncols, nwarps, KQ_stride, use_logit_softcap, needs_fixup, is_fixup>
(Q_f2, K_h2, V_h2, maskh, dstk, dst_meta, scale, slope, logit_softcap,
ne00, ne01, ne02, ne03, ne10, ne11, ne12, ne13, ne31, nb31, nb01, nb02, nb03, nb11, nb12, nb13, nb21, nb22, nb23, ne0, ne1, ne2, ne3,
jt, kb0_start, kb0_stop);
}
kbc += iter_k;
kbc -= kbc % iter_k;
kb0_start = 0;
kb0_stop = min(iter_k, kbc_stop - kbc);
}
if (kbc >= kbc_stop) {
return;
}
const int channel = kbc / (iter_k*iter_j);
const int jt = (kbc - channel*iter_k*iter_j) / iter_k; // j index of current tile.
const float2 * Q_f2 = (const float2 *) (Q + nb02* channel);
const half2 * K_h2 = (const half2 *) (K + nb12*(channel / gqa_ratio));
const half2 * V_h2 = (const half2 *) (V + nb12*(channel / gqa_ratio)); // K and V have same shape
const half * maskh = mask ? (const half *) mask + (nb31/sizeof(half))*jt*ncols : nullptr;
float2 * dstk = ((float2 *) dst) + channel*(D/2);
const float slope = get_alibi_slope(max_bias, channel, n_head_log2, m0, m1);
constexpr bool is_fixup = true; // Last index writes its data to fixup buffer to avoid data races with other blocks.
constexpr bool needs_fixup = false;
flash_attn_ext_f16_process_tile<D, ncols, nwarps, KQ_stride, use_logit_softcap, needs_fixup, is_fixup>
(Q_f2, K_h2, V_h2, maskh, dstk, dst_meta, scale, slope, logit_softcap,
ne00, ne01, ne02, ne03, ne10, ne11, ne12, ne13, ne31, nb31, nb01, nb02, nb03, nb11, nb12, nb13, nb21, nb22, nb23, ne0, ne1, ne2, ne3,
jt, kb0_start, kb0_stop);
}
template <int D, int cols_per_block>
void ggml_cuda_flash_attn_ext_mma_f16_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
typedef mma_A_I16K8<half2> mma_A;
typedef mma_B_J8K8<half2> mma_B;
static_assert(D % mma_B::K == 0, "bad D");
static_assert(cols_per_block % mma_B::J == 0, "bad cols_per_block");
const ggml_tensor * KQV = dst;
constexpr int KQ_stride = D <= 128 ? 64 : 32;
constexpr int nwarps = (KQ_stride == 32 && cols_per_block <= 16) ?
cols_per_block/mma_B::J * KQ_stride/mma_A::I : (cols_per_block <= 8 ? 4 : 8);
constexpr size_t nbytes_shared = std::max(KQ_stride, nwarps*mma_B::J) * (D + 8) * sizeof(half);
float logit_softcap;
memcpy(&logit_softcap, (const float *) KQV->op_params + 2, sizeof(float));
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, KQ_stride, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, KQ_stride, use_logit_softcap>;
}
launch_fattn<D, cols_per_block, 0, KQ_stride>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, true, true);
}
#define DECL_FATTN_MMA_F16_CASE(D, cols_per_block) \
template void ggml_cuda_flash_attn_ext_mma_f16_case \
<D, cols_per_block>(ggml_backend_cuda_context & ctx, ggml_tensor * dst) \
extern DECL_FATTN_MMA_F16_CASE( 64, 8);
extern DECL_FATTN_MMA_F16_CASE( 80, 8);
extern DECL_FATTN_MMA_F16_CASE( 96, 8);
extern DECL_FATTN_MMA_F16_CASE(112, 8);
extern DECL_FATTN_MMA_F16_CASE(128, 8);
extern DECL_FATTN_MMA_F16_CASE(256, 8);
extern DECL_FATTN_MMA_F16_CASE( 64, 16);
extern DECL_FATTN_MMA_F16_CASE( 80, 16);
extern DECL_FATTN_MMA_F16_CASE( 96, 16);
extern DECL_FATTN_MMA_F16_CASE(112, 16);
extern DECL_FATTN_MMA_F16_CASE(128, 16);
extern DECL_FATTN_MMA_F16_CASE(256, 16);
extern DECL_FATTN_MMA_F16_CASE( 64, 32);
extern DECL_FATTN_MMA_F16_CASE( 80, 32);
extern DECL_FATTN_MMA_F16_CASE( 96, 32);
extern DECL_FATTN_MMA_F16_CASE(112, 32);
extern DECL_FATTN_MMA_F16_CASE(128, 32);
extern DECL_FATTN_MMA_F16_CASE(256, 32);
extern DECL_FATTN_MMA_F16_CASE( 64, 64);
extern DECL_FATTN_MMA_F16_CASE( 80, 64);
extern DECL_FATTN_MMA_F16_CASE( 96, 64);
extern DECL_FATTN_MMA_F16_CASE(112, 64);
extern DECL_FATTN_MMA_F16_CASE(128, 64);
extern DECL_FATTN_MMA_F16_CASE(256, 64);

16
ggml/src/ggml-cuda/fattn-tile-f16.cu
View file

@ -45,7 +45,17 @@ static __global__ void flash_attn_tile_ext_f16(
const int ne2,
const int ne3) {
#ifdef FP16_AVAILABLE
#ifndef FLASH_ATTN_AVAILABLE
NO_DEVICE_CODE;
return;
#endif // FLASH_ATTN_AVAILABLE
// Skip unused kernel variants for faster compilation:
#ifdef FP16_MMA_AVAILABLE
NO_DEVICE_CODE;
return;
#endif // FP16_MMA_AVAILABLE
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
return;
@ -290,14 +300,16 @@ void launch_fattn_tile_f16_64_128(ggml_backend_cuda_context & ctx, ggml_tensor *
case 64: {
constexpr int D = 64;
constexpr int nwarps = 8;
constexpr size_t nbytes_shared = 0;
fattn_kernel_t fattn_kernel = flash_attn_tile_ext_f16<D, cols_per_block, nwarps, parallel_blocks, use_logit_softcap>;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, true, true);
} break;
case 128: {
constexpr int D = 128;
constexpr int nwarps = 8;
constexpr size_t nbytes_shared = 0;
fattn_kernel_t fattn_kernel = flash_attn_tile_ext_f16<D, cols_per_block, nwarps, parallel_blocks, use_logit_softcap>;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, true, true);
} break;
default: {
GGML_ABORT("FlashAttention without tensor cores only supports head sizes 64 and 128.");

11
ggml/src/ggml-cuda/fattn-tile-f32.cu
View file

@ -48,7 +48,12 @@ static __global__ void flash_attn_tile_ext_f32(
NO_DEVICE_CODE;
return;
#endif // FLASH_ATTN_AVAILABLE
// Skip unused kernel variants for faster compilation:
#ifdef FP16_MMA_AVAILABLE
NO_DEVICE_CODE;
return;
#endif // FP16_MMA_AVAILABLE
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
return;
@ -289,14 +294,16 @@ void launch_fattn_tile_f32_64_128(ggml_backend_cuda_context & ctx, ggml_tensor *
case 64: {
constexpr int D = 64;
constexpr int nwarps = 8;
constexpr size_t nbytes_shared = 0;
fattn_kernel_t fattn_kernel = flash_attn_tile_ext_f32<D, cols_per_block, nwarps, parallel_blocks, use_logit_softcap>;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, true, true);
} break;
case 128: {
constexpr int D = 128;
constexpr int nwarps = 8;
constexpr size_t nbytes_shared = 0;
fattn_kernel_t fattn_kernel = flash_attn_tile_ext_f32<D, cols_per_block, nwarps, parallel_blocks, use_logit_softcap>;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, true, true);
} break;
default: {
GGML_ABORT("FlashAttention without tensor cores only supports head sizes 64 and 128.");

9
ggml/src/ggml-cuda/fattn-vec-f16.cuh
View file

@ -42,6 +42,12 @@ static __global__ void flash_attn_vec_ext_f16(
const int ne2,
const int ne3) {
#ifdef FP16_AVAILABLE
#ifndef FLASH_ATTN_AVAILABLE
NO_DEVICE_CODE;
return;
#endif // FLASH_ATTN_AVAILABLE
// Skip unused kernel variants for faster compilation:
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
@ -303,7 +309,8 @@ void ggml_cuda_flash_attn_ext_vec_f16_case_impl(ggml_backend_cuda_context & ctx,
fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16<D, cols_per_block, parallel_blocks, type_K, type_V, use_logit_softcap>;
constexpr bool need_f16_K = D != 128;
constexpr bool need_f16_V = D != 128 && D != 64;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V);
constexpr size_t nbytes_shared = 0;
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, need_f16_K, need_f16_V);
}
template <int D, ggml_type type_K, ggml_type type_V>

8
ggml/src/ggml-cuda/fattn-vec-f32.cuh
View file

@ -41,6 +41,11 @@ static __global__ void flash_attn_vec_ext_f32(
const int ne1,
const int ne2,
const int ne3) {
#ifndef FLASH_ATTN_AVAILABLE
NO_DEVICE_CODE;
return;
#endif // FLASH_ATTN_AVAILABLE
// Skip unused kernel variants for faster compilation:
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
@ -284,7 +289,8 @@ void ggml_cuda_flash_attn_ext_vec_f32_case_impl(ggml_backend_cuda_context & ctx,
fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32<D, cols_per_block, parallel_blocks, type_K, type_V, use_logit_softcap>;
constexpr bool need_f16_K = D != 128;
constexpr bool need_f16_V = D != 128 && D != 64;
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, need_f16_K, need_f16_V);
constexpr size_t nbytes_shared = 0;
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, nbytes_shared, need_f16_K, need_f16_V);
}
template <int D, ggml_type type_K, ggml_type type_V>

648
ggml/src/ggml-cuda/fattn-wmma-f16.cu Normal file
View file

@ -0,0 +1,648 @@
// Old and deprecated WMMA FlashAttention implementation.
// It is still needed for Volta since the memory layout of NVIDIA tensor cores changed with Turing.
// Long-term the WMMA code should be replaced with a dedicated Volta implementation.
#include "common.cuh"
#include "fattn-common.cuh"
#include "fattn-wmma-f16.cuh"
#ifdef FP16_MMA_AVAILABLE
#include <mma.h>
#endif // FP16_MMA_AVAILABLE
// D == head size, VKQ_stride == num VKQ rows calculated in parallel:
template<int D, int ncols, int nwarps, int VKQ_stride, int parallel_blocks, typename KQ_acc_t, bool use_logit_softcap>
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
__launch_bounds__(nwarps*WARP_SIZE, 1)
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
static __global__ void flash_attn_ext_f16(
const char * __restrict__ Q,
const char * __restrict__ K,
const char * __restrict__ V,
const char * __restrict__ mask,
float * __restrict__ dst,
float2 * __restrict__ dst_meta,
const float scale,
const float max_bias,
const float m0,
const float m1,
const uint32_t n_head_log2,
const float logit_softcap,
const int ne00,
const int ne01,
const int ne02,
const int ne03,
const int ne10,
const int ne11,
const int ne12,
const int ne13,
const int ne31,
const int nb31,
const int nb01,
const int nb02,
const int nb03,
const int nb11,
const int nb12,
const int nb13,
const int nb21,
const int nb22,
const int nb23,
const int ne0,
const int ne1,
const int ne2,
const int ne3) {
#if __CUDA_ARCH__ == GGML_CUDA_CC_VOLTA
// Skip unused kernel variants for faster compilation:
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
return;
}
//In this kernel Q, K, V are matrices while i, j, k are matrix indices.
const int ic0 = ncols*(blockIdx.x / parallel_blocks); // Index of the first Q/QKV column to work on.
const int ip = blockIdx.x % parallel_blocks; // Index in group of blocks running for the same column in parallel.
static_assert(D <= FATTN_KQ_STRIDE, "D must be <= FATTN_KQ_STRIDE.");
static_assert(ncols == 8 || ncols % 16 == 0, "ncols must be 8 or a multiple of 16.");
constexpr int frag_m = ncols == 8 ? 32 : 16;
constexpr int frag_n = ncols == 8 ? 8 : 16;
static_assert(D % frag_m == 0, "If ncols == 8 then D % frag_m must be 0.");
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a, frag_m, frag_n, 16, half, nvcuda::wmma::row_major> frag_a_K;
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a, frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_a_V;
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_b, frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_b;
typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, KQ_acc_t> frag_c_KQ;
typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, half> frag_c_VKQ;
constexpr int KQ_stride_tc = nwarps*frag_m; // Number of KQ rows calculated in parallel.
constexpr int VKQ_ratio = KQ_stride_tc/VKQ_stride; // Number of parallel VKQ accumulators needed to keep all warps busy.
static_assert(VKQ_ratio <= nwarps, "VKQ_ratio must be <= nwarps.");
// Pad internal representation of KQ, KQV to reduce shared memory bank conflicts:
constexpr int D_padded = D + 8;
constexpr int kqs_padded = FATTN_KQ_STRIDE + 8;
constexpr int kqar = sizeof(KQ_acc_t)/sizeof(half);
const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
const float * Q_f = (const float *) (Q + nb02* blockIdx.y + nb01*ic0);
const half * K_h = (const half *) (K + nb12*(blockIdx.y / gqa_ratio));
const half * V_h = (const half *) (V + nb12*(blockIdx.y / gqa_ratio)); // K and V have same shape
const half * maskh = (const half *) mask + (nb31/sizeof(half))* ic0;
const half2 * mask2 = (const half2 *) mask + (nb31/sizeof(half))*(ic0/2);
const int stride_Q = nb01 / sizeof(float);
const int stride_KV = nb11 / sizeof(half);
const float slopef = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
const half slopeh = __float2half(slopef);
const half2 slope2 = make_half2(slopef, slopef);
const half2 logit_softcap_2 = make_half2(logit_softcap, logit_softcap);
frag_b Q_b[D/16][ncols/frag_n];
// A single buffer for temporarily holding tiles of KQ and VKQ parts:
constexpr int mem_KQ = ncols*kqs_padded*kqar;
constexpr int mem_VKQ_parts = VKQ_ratio*ncols*D_padded;
__shared__ half KQ[mem_KQ >= mem_VKQ_parts ? mem_KQ : mem_VKQ_parts];
float * KQ_f = (float *) KQ;
half2 * KQ2 = (half2 *) KQ;
float KQ_rowsum_f[ncols/nwarps] = {0.0f};
float KQ_max_f[ncols/nwarps];
float KQ_max_scale_f[ncols/nwarps] = {0.0f};
#pragma unroll
for (int j = 0; j < ncols/nwarps; ++j) {
KQ_max_f[j] = -FLT_MAX/2.0f;
}
half2 KQ_rowsum_h2[ncols/nwarps] = {{0.0f, 0.0f}};
half2 KQ_max_h2[ncols/nwarps];
half2 KQ_max_scale_h2[ncols/nwarps] = {{0.0f, 0.0f}};
#pragma unroll
for (int j = 0; j < ncols/nwarps; ++j) {
KQ_max_h2[j] = make_half2(-HALF_MAX_HALF, -HALF_MAX_HALF);
}
__shared__ half VKQ[ncols*D_padded]; // Accumulator for final VKQ slice.
half2 * VKQ2 = (half2 *) VKQ;
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
#pragma unroll
for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D/2 && i >= D/2) {
break;
}
VKQ2[j*(D_padded/2) + i] = make_half2(0.0f, 0.0f);
}
}
// Convert Q to half and apply scale, temporarily store in KQ:
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
#pragma unroll
for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D && i >= D) {
break;
}
KQ[j*D_padded + i] = ic0 + j < ne01 ? Q_f[j*stride_Q + i] * scale : 0.0f;
}
}
__syncthreads();
// Load Q into tensor core fragments/registers since it will be used frequently:
#pragma unroll
for (int i0 = 0; i0 < D; i0 += 16) {
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::load_matrix_sync(Q_b[i0/16][j0/frag_n], KQ + j0*D_padded + i0, D_padded);
}
}
__syncthreads();
// Iterate over ne11 == previous tokens:
for (int k_VKQ_0 = ip*FATTN_KQ_STRIDE; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*FATTN_KQ_STRIDE) {
// Calculate tile of KQ:
#pragma unroll
for (int i_KQ_0 = 0; i_KQ_0 < FATTN_KQ_STRIDE; i_KQ_0 += KQ_stride_tc) {
frag_c_KQ KQ_c[ncols/frag_n];
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::fill_fragment(KQ_c[j], 0.0f);
}
#pragma unroll
for (int k_KQ_0 = 0; k_KQ_0 < D; k_KQ_0 += 16) {
frag_a_K K_a;
nvcuda::wmma::load_matrix_sync(K_a, K_h + (k_VKQ_0 + i_KQ_0 + frag_m*threadIdx.y)*stride_KV + k_KQ_0, stride_KV);
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::mma_sync(KQ_c[j], K_a, Q_b[k_KQ_0/16][j], KQ_c[j]);
}
}
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::store_matrix_sync((KQ_acc_t *) KQ + j0*kqs_padded + i_KQ_0 + frag_m*threadIdx.y, KQ_c[j0/frag_n], kqs_padded, nvcuda::wmma::mem_col_major);
}
}
__syncthreads();
// Calculate softmax for each KQ column using the current max. value.
// The divisor is stored in KQ_rowsum and will be applied at the end.
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
if (std::is_same<KQ_acc_t, float>::value) {
float KQ_f_tmp[FATTN_KQ_STRIDE / WARP_SIZE];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ_f_tmp[k0/WARP_SIZE] = KQ_f[j*kqs_padded + k];
if (use_logit_softcap) {
KQ_f_tmp[k0/WARP_SIZE] = logit_softcap*tanhf(KQ_f_tmp[k0/WARP_SIZE]);
}
}
float KQ_max_new = KQ_max_f[j0/nwarps];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ_f_tmp[k0/WARP_SIZE] += mask ? __half2float(slopeh*maskh[j*(nb31/sizeof(half)) + k_VKQ_0 + k]) : 0.0f;
KQ_max_new = max(KQ_max_new, KQ_f_tmp[k0/WARP_SIZE]);
}
KQ_max_new = warp_reduce_max(KQ_max_new);
const float diff = KQ_max_f[j0/nwarps] - KQ_max_new;
KQ_max_scale_f[j0/nwarps] = expf(diff);
if (diff <= SOFTMAX_FTZ_THRESHOLD) {
KQ_max_scale_f[j0/nwarps] = 0.0f;
}
KQ_max_f[j0/nwarps] = KQ_max_new;
float KQ_rowsum_add = 0.0f;
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
const float diff = KQ_f_tmp[k0/WARP_SIZE] - KQ_max_f[j0/nwarps];
KQ_f_tmp[k0/WARP_SIZE] = expf(diff);
if (diff <= SOFTMAX_FTZ_THRESHOLD) {
KQ_f_tmp[k0/WARP_SIZE] = 0.0f;
}
KQ_rowsum_add += KQ_f_tmp[k0/WARP_SIZE];
KQ[j*(kqar*kqs_padded) + k] = KQ_f_tmp[k0/WARP_SIZE];
}
KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
// Scale previous KQ_rowsum to account for a potential increase in KQ_max:
KQ_rowsum_f[j0/nwarps] = KQ_max_scale_f[j0/nwarps]*KQ_rowsum_f[j0/nwarps] + KQ_rowsum_add;
} else {
half2 KQ2_tmp[FATTN_KQ_STRIDE/(2*WARP_SIZE)];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ2_tmp[k0/WARP_SIZE] = KQ2[j*(kqs_padded/2) + k];
if (use_logit_softcap) {
// There is no dedicated tangens hyperbolicus function for half2.
KQ2_tmp[k0/WARP_SIZE] = h2exp(KQ2_tmp[k0/WARP_SIZE]*make_half2(2.0f, 2.0f));
KQ2_tmp[k0/WARP_SIZE] = (KQ2_tmp[k0/WARP_SIZE] - make_half2(1.0f, 1.0f))
/(KQ2_tmp[k0/WARP_SIZE] + make_half2(1.0f, 1.0f));
KQ2_tmp[k0/WARP_SIZE] *= logit_softcap_2;
}
}
half2 KQ_max_new = KQ_max_h2[j0/nwarps];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ2_tmp[k0/WARP_SIZE] += mask ? slope2*mask2[(j*ne11 + k_VKQ_0)/2 + k] : make_half2(0.0f, 0.0f);
KQ_max_new = ggml_cuda_hmax2(KQ_max_new, KQ2_tmp[k0/WARP_SIZE]);
}
KQ_max_new = __half2half2(warp_reduce_max(ggml_cuda_hmax(__low2half(KQ_max_new), __high2half(KQ_max_new))));
const half2 diff = KQ_max_h2[j0/nwarps] - KQ_max_new;
KQ_max_scale_h2[j0/nwarps] = h2exp(diff);
const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
*((uint32_t *) &KQ_max_scale_h2[j0/nwarps]) &= ftz_mask;
KQ_max_h2[j0/nwarps] = KQ_max_new;
half2 KQ_rowsum_add = make_half2(0.0f, 0.0f);
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
const half2 diff = KQ2_tmp[k0/WARP_SIZE] - KQ_max_h2[j0/nwarps];
KQ2_tmp[k0/WARP_SIZE] = h2exp(diff);
const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
*((uint32_t *) &KQ2_tmp[k0/WARP_SIZE]) &= ftz_mask;
KQ_rowsum_add += KQ2_tmp[k0/WARP_SIZE];
KQ2[j*(kqs_padded/2) + k] = KQ2_tmp[k0/WARP_SIZE];
}
KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
// Scale previous KQ_rowsum to account for a potential increase in KQ_max:
KQ_rowsum_h2[j0/nwarps] = KQ_max_scale_h2[j0/nwarps]*KQ_rowsum_h2[j0/nwarps] + KQ_rowsum_add;
}
}
__syncthreads();
frag_b KQ_b[FATTN_KQ_STRIDE/(VKQ_ratio*16)][ncols/frag_n];
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
nvcuda::wmma::load_matrix_sync(
KQ_b[k0/(VKQ_ratio*16)][j0/frag_n],
KQ + j0*(kqar*kqs_padded) + k,
kqar*kqs_padded);
}
}
frag_c_VKQ VKQ_c[D/VKQ_stride][ncols/frag_n];
#pragma unroll
for (int i_VKQ_0 = 0; i_VKQ_0 < D; i_VKQ_0 += VKQ_stride) {
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::fill_fragment(VKQ_c[i_VKQ_0/VKQ_stride][j], 0.0f);
}
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
frag_a_V v_a;
nvcuda::wmma::load_matrix_sync(v_a, V_h + (k_VKQ_0 + k)*stride_KV + i_VKQ_0 + frag_m*(threadIdx.y/VKQ_ratio), stride_KV);
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::mma_sync(VKQ_c[i_VKQ_0/VKQ_stride][j], v_a, KQ_b[k0/(VKQ_ratio*16)][j], VKQ_c[i_VKQ_0/VKQ_stride][j]);
}
}
}
__syncthreads();
const int offset_k = (threadIdx.y % VKQ_ratio) * (ncols*D_padded);
#pragma unroll
for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += VKQ_stride) {
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::store_matrix_sync(
KQ + offset_k + j0*D_padded + i_KQ_0 + frag_m*(threadIdx.y/VKQ_ratio),
VKQ_c[i_KQ_0/VKQ_stride][j0/frag_n],
D_padded, nvcuda::wmma::mem_col_major);
}
}
__syncthreads();
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
half2 VKQ_scale;
if (std::is_same<KQ_acc_t, float>::value) {
VKQ_scale = make_half2(KQ_max_scale_f[j0/nwarps], KQ_max_scale_f[j0/nwarps]);
} else {
VKQ_scale = KQ_max_scale_h2[j0/nwarps];
}
#pragma unroll
for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D/2 && i >= D/2) {
break;
}
half2 VKQ_add = make_half2(0.0f, 0.0f);
#pragma unroll
for (int l = 0; l < VKQ_ratio; ++l) {
VKQ_add += KQ2[l*(ncols*D_padded/2) + j*(D_padded/2) + i];
}
VKQ2[j*(D_padded/2) + i] = VKQ_scale*VKQ2[j*(D_padded/2) + i] + VKQ_add;
}
}
__syncthreads();
}
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j_VKQ = j0 + threadIdx.y;
if (ic0 + j_VKQ >= ne01) {
return;
}
const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
float KQ_rowsum_j;
if (std::is_same<KQ_acc_t, float>::value) {
KQ_rowsum_j = KQ_rowsum_f[j0/nwarps];
} else {
KQ_rowsum_j = __low2float(KQ_rowsum_h2[j0/nwarps]) + __high2float(KQ_rowsum_h2[j0/nwarps]);
}
#pragma unroll
for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D && i >= D) {
break;
}
float dst_val = VKQ[j_VKQ*D_padded + i];
if (parallel_blocks == 1) {
dst_val /= KQ_rowsum_j;
}
dst[j_dst*gridDim.y*D + blockIdx.y*D + i] = dst_val;
}
if (parallel_blocks == 1 || threadIdx.x != 0) {
continue;
}
float2 dst_meta_val;
if (std::is_same<KQ_acc_t, float>::value) {
dst_meta_val.x = KQ_max_f[j0/nwarps];
} else {
dst_meta_val.x = __low2float(KQ_max_h2[j0/nwarps]);
}
dst_meta_val.y = KQ_rowsum_j;
dst_meta[(ic0 + j_VKQ)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = dst_meta_val;
}
#else
NO_DEVICE_CODE;
#endif // __CUDA_ARCH__ == GGML_CUDA_CC_VOLTA
}
constexpr int get_max_power_of_2(int x) {
return x % 2 == 0 ? 2*get_max_power_of_2(x/2) : 1;
}
static_assert(get_max_power_of_2(1) == 1, "Test failed.");
static_assert(get_max_power_of_2(2) == 2, "Test failed.");
static_assert(get_max_power_of_2(4) == 4, "Test failed.");
static_assert(get_max_power_of_2(6) == 2, "Test failed.");
// Number of VKQ rows calculated in parallel:
constexpr int get_VKQ_stride(int D, int nwarps, int frag_m) {
return (get_max_power_of_2(D/frag_m) < nwarps ? get_max_power_of_2(D/frag_m) : nwarps)*frag_m;
}
static_assert(get_VKQ_stride(128, 1, 32) == 32, "Test failed.");
static_assert(get_VKQ_stride(128, 2, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride(128, 4, 32) == 128, "Test failed.");
static_assert(get_VKQ_stride( 64, 1, 32) == 32, "Test failed.");
static_assert(get_VKQ_stride( 64, 2, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride( 64, 4, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride( 80, 1, 16) == 16, "Test failed.");
static_assert(get_VKQ_stride( 80, 2, 16) == 16, "Test failed.");
static_assert(get_VKQ_stride( 80, 4, 16) == 16, "Test failed.");
template <int D, int cols_per_block, typename KQ_acc_t>
void ggml_cuda_flash_attn_ext_wmma_f16_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * KQV = dst;
const ggml_tensor * Q = dst->src[0];
constexpr int nwarps = 4;
constexpr int frag_m = cols_per_block == 8 && D % 32 == 0 ? 32 : 16;
const int blocks_num_pb1 = ((Q->ne[1] + cols_per_block - 1) / cols_per_block)*Q->ne[2]*Q->ne[3];
const int nsm = ggml_cuda_info().devices[ggml_cuda_get_device()].nsm;
float logit_softcap;
memcpy(&logit_softcap, (const float *) KQV->op_params + 2, sizeof(float));
if (4*blocks_num_pb1 < 2*nsm) {
constexpr int parallel_blocks = 4;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, 0, true, true);
return;
}
if (2*blocks_num_pb1 < 2*nsm) {
constexpr int parallel_blocks = 2;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, 0, true, true);
return;
}
constexpr int parallel_blocks = 1;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, cols_per_block, parallel_blocks, -1>(ctx, dst, fattn_kernel, nwarps, 0, true, true);
}
void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * KQV = dst;
const ggml_tensor * Q = dst->src[0];
const enum ggml_prec prec = ggml_flash_attn_ext_get_prec(KQV);
if (prec != GGML_PREC_DEFAULT) {
if (Q->ne[1] <= 32 || Q->ne[0] > 128) {
constexpr int cols_per_block = 16;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, float>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
} else {
constexpr int cols_per_block = 32;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
break;
// case 256:
// ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, float>(ctx, dst);
// break;
default:
GGML_ABORT("fatal error");
break;
}
}
return;
}
if (Q->ne[1] <= 8 && Q->ne[0] % WARP_SIZE == 0) {
constexpr int cols_per_block = 8;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
return;
}
if (Q->ne[1] <= 32) {
constexpr int cols_per_block = 16;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
return;
}
constexpr int cols_per_block = 32;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
}

542
ggml/src/ggml-cuda/fattn-wmma-f16.cuh
View file

@ -1,543 +1,3 @@
#include "common.cuh"
#include "fattn-common.cuh"
#ifdef FP16_MMA_AVAILABLE
#include <mma.h>
#endif // FP16_MMA_AVAILABLE
// D == head size, VKQ_stride == num VKQ rows calculated in parallel:
template<int D, int ncols, int nwarps, int VKQ_stride, int parallel_blocks, typename KQ_acc_t, bool use_logit_softcap>
#if !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
__launch_bounds__(nwarps*WARP_SIZE, 1)
#endif // !(defined(GGML_USE_HIP) && defined(__HIP_PLATFORM_AMD__))
static __global__ void flash_attn_ext_f16(
const char * __restrict__ Q,
const char * __restrict__ K,
const char * __restrict__ V,
const char * __restrict__ mask,
float * __restrict__ dst,
float2 * __restrict__ dst_meta,
const float scale,
const float max_bias,
const float m0,
const float m1,
const uint32_t n_head_log2,
const float logit_softcap,
const int ne00,
const int ne01,
const int ne02,
const int ne03,
const int ne10,
const int ne11,
const int ne12,
const int ne13,
const int ne31,
const int nb31,
const int nb01,
const int nb02,
const int nb03,
const int nb11,
const int nb12,
const int nb13,
const int nb21,
const int nb22,
const int nb23,
const int ne0,
const int ne1,
const int ne2,
const int ne3) {
#ifdef FP16_MMA_AVAILABLE
// Skip unused kernel variants for faster compilation:
if (use_logit_softcap && !(D == 128 || D == 256)) {
NO_DEVICE_CODE;
return;
}
//In this kernel Q, K, V are matrices while i, j, k are matrix indices.
const int ic0 = ncols*(blockIdx.x / parallel_blocks); // Index of the first Q/QKV column to work on.
const int ip = blockIdx.x % parallel_blocks; // Index in group of blocks running for the same column in parallel.
static_assert(D <= FATTN_KQ_STRIDE, "D must be <= FATTN_KQ_STRIDE.");
static_assert(ncols == 8 || ncols % 16 == 0, "ncols must be 8 or a multiple of 16.");
constexpr int frag_m = ncols == 8 ? 32 : 16;
constexpr int frag_n = ncols == 8 ? 8 : 16;
static_assert(D % frag_m == 0, "If ncols == 8 then D % frag_m must be 0.");
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a, frag_m, frag_n, 16, half, nvcuda::wmma::row_major> frag_a_K;
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a, frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_a_V;
typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_b, frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_b;
typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, KQ_acc_t> frag_c_KQ;
typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, half> frag_c_VKQ;
constexpr int KQ_stride_tc = nwarps*frag_m; // Number of KQ rows calculated in parallel.
constexpr int VKQ_ratio = KQ_stride_tc/VKQ_stride; // Number of parallel VKQ accumulators needed to keep all warps busy.
static_assert(VKQ_ratio <= nwarps, "VKQ_ratio must be <= nwarps.");
// Pad internal representation of KQ, KQV to reduce shared memory bank conflicts:
constexpr int D_padded = D + 8;
constexpr int kqs_padded = FATTN_KQ_STRIDE + 8;
constexpr int kqar = sizeof(KQ_acc_t)/sizeof(half);
const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
const float * Q_f = (const float *) (Q + nb02* blockIdx.y + nb01*ic0);
const half * K_h = (const half *) (K + nb12*(blockIdx.y / gqa_ratio));
const half * V_h = (const half *) (V + nb12*(blockIdx.y / gqa_ratio)); // K and V have same shape
const half * maskh = (const half *) mask + (nb31/sizeof(half))* ic0;
const half2 * mask2 = (const half2 *) mask + (nb31/sizeof(half))*(ic0/2);
const int stride_Q = nb01 / sizeof(float);
const int stride_KV = nb11 / sizeof(half);
const float slopef = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
const half slopeh = __float2half(slopef);
const half2 slope2 = make_half2(slopef, slopef);
const half2 logit_softcap_2 = make_half2(logit_softcap, logit_softcap);
frag_b Q_b[D/16][ncols/frag_n];
// A single buffer for temporarily holding tiles of KQ and VKQ parts:
constexpr int mem_KQ = ncols*kqs_padded*kqar;
constexpr int mem_VKQ_parts = VKQ_ratio*ncols*D_padded;
__shared__ half KQ[mem_KQ >= mem_VKQ_parts ? mem_KQ : mem_VKQ_parts];
float * KQ_f = (float *) KQ;
half2 * KQ2 = (half2 *) KQ;
float KQ_rowsum_f[ncols/nwarps] = {0.0f};
float KQ_max_f[ncols/nwarps];
float KQ_max_scale_f[ncols/nwarps] = {0.0f};
#pragma unroll
for (int j = 0; j < ncols/nwarps; ++j) {
KQ_max_f[j] = -FLT_MAX/2.0f;
}
half2 KQ_rowsum_h2[ncols/nwarps] = {{0.0f, 0.0f}};
half2 KQ_max_h2[ncols/nwarps];
half2 KQ_max_scale_h2[ncols/nwarps] = {{0.0f, 0.0f}};
#pragma unroll
for (int j = 0; j < ncols/nwarps; ++j) {
KQ_max_h2[j] = make_half2(-HALF_MAX_HALF, -HALF_MAX_HALF);
}
__shared__ half VKQ[ncols*D_padded]; // Accumulator for final VKQ slice.
half2 * VKQ2 = (half2 *) VKQ;
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
#pragma unroll
for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D/2 && i >= D/2) {
break;
}
VKQ2[j*(D_padded/2) + i] = make_half2(0.0f, 0.0f);
}
}
// Convert Q to half and apply scale, temporarily store in KQ:
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
#pragma unroll
for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D && i >= D) {
break;
}
KQ[j*D_padded + i] = ic0 + j < ne01 ? Q_f[j*stride_Q + i] * scale : 0.0f;
}
}
__syncthreads();
// Load Q into tensor core fragments/registers since it will be used frequently:
#pragma unroll
for (int i0 = 0; i0 < D; i0 += 16) {
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::load_matrix_sync(Q_b[i0/16][j0/frag_n], KQ + j0*D_padded + i0, D_padded);
}
}
__syncthreads();
// Iterate over ne11 == previous tokens:
for (int k_VKQ_0 = ip*FATTN_KQ_STRIDE; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*FATTN_KQ_STRIDE) {
// Calculate tile of KQ:
#pragma unroll
for (int i_KQ_0 = 0; i_KQ_0 < FATTN_KQ_STRIDE; i_KQ_0 += KQ_stride_tc) {
frag_c_KQ KQ_c[ncols/frag_n];
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::fill_fragment(KQ_c[j], 0.0f);
}
#pragma unroll
for (int k_KQ_0 = 0; k_KQ_0 < D; k_KQ_0 += 16) {
frag_a_K K_a;
nvcuda::wmma::load_matrix_sync(K_a, K_h + (k_VKQ_0 + i_KQ_0 + frag_m*threadIdx.y)*stride_KV + k_KQ_0, stride_KV);
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::mma_sync(KQ_c[j], K_a, Q_b[k_KQ_0/16][j], KQ_c[j]);
}
}
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::store_matrix_sync((KQ_acc_t *) KQ + j0*kqs_padded + i_KQ_0 + frag_m*threadIdx.y, KQ_c[j0/frag_n], kqs_padded, nvcuda::wmma::mem_col_major);
}
}
__syncthreads();
// Calculate softmax for each KQ column using the current max. value.
// The divisor is stored in KQ_rowsum and will be applied at the end.
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
if (std::is_same<KQ_acc_t, float>::value) {
float KQ_f_tmp[FATTN_KQ_STRIDE / WARP_SIZE];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ_f_tmp[k0/WARP_SIZE] = KQ_f[j*kqs_padded + k];
if (use_logit_softcap) {
KQ_f_tmp[k0/WARP_SIZE] = logit_softcap*tanhf(KQ_f_tmp[k0/WARP_SIZE]);
}
}
float KQ_max_new = KQ_max_f[j0/nwarps];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ_f_tmp[k0/WARP_SIZE] += mask ? __half2float(slopeh*maskh[j*(nb31/sizeof(half)) + k_VKQ_0 + k]) : 0.0f;
KQ_max_new = max(KQ_max_new, KQ_f_tmp[k0/WARP_SIZE]);
}
KQ_max_new = warp_reduce_max(KQ_max_new);
const float diff = KQ_max_f[j0/nwarps] - KQ_max_new;
KQ_max_scale_f[j0/nwarps] = expf(diff);
if (diff <= SOFTMAX_FTZ_THRESHOLD) {
KQ_max_scale_f[j0/nwarps] = 0.0f;
}
KQ_max_f[j0/nwarps] = KQ_max_new;
float KQ_rowsum_add = 0.0f;
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
const float diff = KQ_f_tmp[k0/WARP_SIZE] - KQ_max_f[j0/nwarps];
KQ_f_tmp[k0/WARP_SIZE] = expf(diff);
if (diff <= SOFTMAX_FTZ_THRESHOLD) {
KQ_f_tmp[k0/WARP_SIZE] = 0.0f;
}
KQ_rowsum_add += KQ_f_tmp[k0/WARP_SIZE];
KQ[j*(kqar*kqs_padded) + k] = KQ_f_tmp[k0/WARP_SIZE];
}
KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
// Scale previous KQ_rowsum to account for a potential increase in KQ_max:
KQ_rowsum_f[j0/nwarps] = KQ_max_scale_f[j0/nwarps]*KQ_rowsum_f[j0/nwarps] + KQ_rowsum_add;
} else {
half2 KQ2_tmp[FATTN_KQ_STRIDE/(2*WARP_SIZE)];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ2_tmp[k0/WARP_SIZE] = KQ2[j*(kqs_padded/2) + k];
if (use_logit_softcap) {
// There is no dedicated tangens hyperbolicus function for half2.
KQ2_tmp[k0/WARP_SIZE] = h2exp(KQ2_tmp[k0/WARP_SIZE]*make_half2(2.0f, 2.0f));
KQ2_tmp[k0/WARP_SIZE] = (KQ2_tmp[k0/WARP_SIZE] - make_half2(1.0f, 1.0f))
/(KQ2_tmp[k0/WARP_SIZE] + make_half2(1.0f, 1.0f));
KQ2_tmp[k0/WARP_SIZE] *= logit_softcap_2;
}
}
half2 KQ_max_new = KQ_max_h2[j0/nwarps];
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
KQ2_tmp[k0/WARP_SIZE] += mask ? slope2*mask2[(j*ne11 + k_VKQ_0)/2 + k] : make_half2(0.0f, 0.0f);
KQ_max_new = ggml_cuda_hmax2(KQ_max_new, KQ2_tmp[k0/WARP_SIZE]);
}
KQ_max_new = __half2half2(warp_reduce_max(ggml_cuda_hmax(__low2half(KQ_max_new), __high2half(KQ_max_new))));
const half2 diff = KQ_max_h2[j0/nwarps] - KQ_max_new;
KQ_max_scale_h2[j0/nwarps] = h2exp(diff);
const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
*((uint32_t *) &KQ_max_scale_h2[j0/nwarps]) &= ftz_mask;
KQ_max_h2[j0/nwarps] = KQ_max_new;
half2 KQ_rowsum_add = make_half2(0.0f, 0.0f);
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
const int k = k0 + threadIdx.x;
const half2 diff = KQ2_tmp[k0/WARP_SIZE] - KQ_max_h2[j0/nwarps];
KQ2_tmp[k0/WARP_SIZE] = h2exp(diff);
const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
*((uint32_t *) &KQ2_tmp[k0/WARP_SIZE]) &= ftz_mask;
KQ_rowsum_add += KQ2_tmp[k0/WARP_SIZE];
KQ2[j*(kqs_padded/2) + k] = KQ2_tmp[k0/WARP_SIZE];
}
KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
// Scale previous KQ_rowsum to account for a potential increase in KQ_max:
KQ_rowsum_h2[j0/nwarps] = KQ_max_scale_h2[j0/nwarps]*KQ_rowsum_h2[j0/nwarps] + KQ_rowsum_add;
}
}
__syncthreads();
frag_b KQ_b[FATTN_KQ_STRIDE/(VKQ_ratio*16)][ncols/frag_n];
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
nvcuda::wmma::load_matrix_sync(
KQ_b[k0/(VKQ_ratio*16)][j0/frag_n],
KQ + j0*(kqar*kqs_padded) + k,
kqar*kqs_padded);
}
}
frag_c_VKQ VKQ_c[D/VKQ_stride][ncols/frag_n];
#pragma unroll
for (int i_VKQ_0 = 0; i_VKQ_0 < D; i_VKQ_0 += VKQ_stride) {
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::fill_fragment(VKQ_c[i_VKQ_0/VKQ_stride][j], 0.0f);
}
#pragma unroll
for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
frag_a_V v_a;
nvcuda::wmma::load_matrix_sync(v_a, V_h + (k_VKQ_0 + k)*stride_KV + i_VKQ_0 + frag_m*(threadIdx.y/VKQ_ratio), stride_KV);
#pragma unroll
for (int j = 0; j < ncols/frag_n; ++j) {
nvcuda::wmma::mma_sync(VKQ_c[i_VKQ_0/VKQ_stride][j], v_a, KQ_b[k0/(VKQ_ratio*16)][j], VKQ_c[i_VKQ_0/VKQ_stride][j]);
}
}
}
__syncthreads();
const int offset_k = (threadIdx.y % VKQ_ratio) * (ncols*D_padded);
#pragma unroll
for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += VKQ_stride) {
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += frag_n) {
nvcuda::wmma::store_matrix_sync(
KQ + offset_k + j0*D_padded + i_KQ_0 + frag_m*(threadIdx.y/VKQ_ratio),
VKQ_c[i_KQ_0/VKQ_stride][j0/frag_n],
D_padded, nvcuda::wmma::mem_col_major);
}
}
__syncthreads();
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j = j0 + threadIdx.y;
half2 VKQ_scale;
if (std::is_same<KQ_acc_t, float>::value) {
VKQ_scale = make_half2(KQ_max_scale_f[j0/nwarps], KQ_max_scale_f[j0/nwarps]);
} else {
VKQ_scale = KQ_max_scale_h2[j0/nwarps];
}
#pragma unroll
for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D/2 && i >= D/2) {
break;
}
half2 VKQ_add = make_half2(0.0f, 0.0f);
#pragma unroll
for (int l = 0; l < VKQ_ratio; ++l) {
VKQ_add += KQ2[l*(ncols*D_padded/2) + j*(D_padded/2) + i];
}
VKQ2[j*(D_padded/2) + i] = VKQ_scale*VKQ2[j*(D_padded/2) + i] + VKQ_add;
}
}
__syncthreads();
}
#pragma unroll
for (int j0 = 0; j0 < ncols; j0 += nwarps) {
const int j_VKQ = j0 + threadIdx.y;
if (ic0 + j_VKQ >= ne01) {
return;
}
const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
float KQ_rowsum_j;
if (std::is_same<KQ_acc_t, float>::value) {
KQ_rowsum_j = KQ_rowsum_f[j0/nwarps];
} else {
KQ_rowsum_j = __low2float(KQ_rowsum_h2[j0/nwarps]) + __high2float(KQ_rowsum_h2[j0/nwarps]);
}
#pragma unroll
for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
const int i = i0 + threadIdx.x;
if (i0 + WARP_SIZE > D && i >= D) {
break;
}
float dst_val = VKQ[j_VKQ*D_padded + i];
if (parallel_blocks == 1) {
dst_val /= KQ_rowsum_j;
}
dst[j_dst*gridDim.y*D + blockIdx.y*D + i] = dst_val;
}
if (parallel_blocks == 1 || threadIdx.x != 0) {
continue;
}
float2 dst_meta_val;
if (std::is_same<KQ_acc_t, float>::value) {
dst_meta_val.x = KQ_max_f[j0/nwarps];
} else {
dst_meta_val.x = __low2float(KQ_max_h2[j0/nwarps]);
}
dst_meta_val.y = KQ_rowsum_j;
dst_meta[(ic0 + j_VKQ)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = dst_meta_val;
}
#else
NO_DEVICE_CODE;
#endif // FP16_MMA_AVAILABLE
}
constexpr int get_max_power_of_2(int x) {
return x % 2 == 0 ? 2*get_max_power_of_2(x/2) : 1;
}
static_assert(get_max_power_of_2(1) == 1, "Test failed.");
static_assert(get_max_power_of_2(2) == 2, "Test failed.");
static_assert(get_max_power_of_2(4) == 4, "Test failed.");
static_assert(get_max_power_of_2(6) == 2, "Test failed.");
// Number of VKQ rows calculated in parallel:
constexpr int get_VKQ_stride(int D, int nwarps, int frag_m) {
return (get_max_power_of_2(D/frag_m) < nwarps ? get_max_power_of_2(D/frag_m) : nwarps)*frag_m;
}
static_assert(get_VKQ_stride(128, 1, 32) == 32, "Test failed.");
static_assert(get_VKQ_stride(128, 2, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride(128, 4, 32) == 128, "Test failed.");
static_assert(get_VKQ_stride( 64, 1, 32) == 32, "Test failed.");
static_assert(get_VKQ_stride( 64, 2, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride( 64, 4, 32) == 64, "Test failed.");
static_assert(get_VKQ_stride( 80, 1, 16) == 16, "Test failed.");
static_assert(get_VKQ_stride( 80, 2, 16) == 16, "Test failed.");
static_assert(get_VKQ_stride( 80, 4, 16) == 16, "Test failed.");
template <int D, int cols_per_block, typename KQ_acc_t>
void ggml_cuda_flash_attn_ext_wmma_f16_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * KQV = dst;
const ggml_tensor * Q = dst->src[0];
constexpr int nwarps = 4;
constexpr int frag_m = cols_per_block == 8 && D % 32 == 0 ? 32 : 16;
const int blocks_num_pb1 = ((Q->ne[1] + cols_per_block - 1) / cols_per_block)*Q->ne[2]*Q->ne[3];
const int nsm = ggml_cuda_info().devices[ggml_cuda_get_device()].nsm;
float logit_softcap;
memcpy(&logit_softcap, (const float *) KQV->op_params + 2, sizeof(float));
if (4*blocks_num_pb1 < 2*nsm) {
constexpr int parallel_blocks = 4;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
return;
}
if (2*blocks_num_pb1 < 2*nsm) {
constexpr int parallel_blocks = 2;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
return;
}
constexpr int parallel_blocks = 1;
fattn_kernel_t fattn_kernel;
if (logit_softcap == 0.0f) {
constexpr bool use_logit_softcap = false;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
} else {
constexpr bool use_logit_softcap = true;
fattn_kernel = flash_attn_ext_f16<
D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t, use_logit_softcap>;
}
launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block, true, true);
}
#define DECL_FATTN_WMMA_F16_CASE(D, cols_per_block, KQ_acc_t) \
template void ggml_cuda_flash_attn_ext_wmma_f16_case \
<D, cols_per_block, KQ_acc_t>(ggml_backend_cuda_context & ctx, ggml_tensor * dst) \
extern DECL_FATTN_WMMA_F16_CASE( 64, 16, float);
extern DECL_FATTN_WMMA_F16_CASE( 80, 16, float);
extern DECL_FATTN_WMMA_F16_CASE( 96, 16, float);
extern DECL_FATTN_WMMA_F16_CASE(112, 16, float);
extern DECL_FATTN_WMMA_F16_CASE(128, 16, float);
extern DECL_FATTN_WMMA_F16_CASE(256, 16, float);
extern DECL_FATTN_WMMA_F16_CASE( 64, 32, float);
extern DECL_FATTN_WMMA_F16_CASE( 80, 32, float);
extern DECL_FATTN_WMMA_F16_CASE( 96, 32, float);
extern DECL_FATTN_WMMA_F16_CASE(112, 32, float);
extern DECL_FATTN_WMMA_F16_CASE(128, 32, float);
// extern DECL_FATTN_WMMA_F16_CASE(256, 16, float);
extern DECL_FATTN_WMMA_F16_CASE( 64, 8, half);
extern DECL_FATTN_WMMA_F16_CASE( 96, 8, half);
extern DECL_FATTN_WMMA_F16_CASE(128, 8, half);
extern DECL_FATTN_WMMA_F16_CASE(256, 8, half);
extern DECL_FATTN_WMMA_F16_CASE( 64, 16, half);
extern DECL_FATTN_WMMA_F16_CASE( 80, 16, half);
extern DECL_FATTN_WMMA_F16_CASE( 96, 16, half);
extern DECL_FATTN_WMMA_F16_CASE(112, 16, half);
extern DECL_FATTN_WMMA_F16_CASE(128, 16, half);
extern DECL_FATTN_WMMA_F16_CASE(256, 16, half);
extern DECL_FATTN_WMMA_F16_CASE( 64, 32, half);
extern DECL_FATTN_WMMA_F16_CASE( 80, 32, half);
extern DECL_FATTN_WMMA_F16_CASE( 96, 32, half);
extern DECL_FATTN_WMMA_F16_CASE(112, 32, half);
extern DECL_FATTN_WMMA_F16_CASE(128, 32, half);
extern DECL_FATTN_WMMA_F16_CASE(256, 16, half);
void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst);

141
ggml/src/ggml-cuda/fattn.cu
View file

@ -1,5 +1,6 @@
#include "common.cuh"
#include "fattn-common.cuh"
#include "fattn-mma-f16.cuh"
#include "fattn-tile-f16.cuh"
#include "fattn-tile-f32.cuh"
#include "fattn-vec-f16.cuh"
@ -7,144 +8,56 @@
#include "fattn-wmma-f16.cuh"
#include "fattn.cuh"
#include <cstdint>
static void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * KQV = dst;
template <int cols_per_block>
static void ggml_cuda_flash_attn_ext_mma_f16_switch_hs(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * Q = dst->src[0];
const enum ggml_prec prec = ggml_flash_attn_ext_get_prec(KQV);
if (prec != GGML_PREC_DEFAULT) {
if (Q->ne[1] <= 32 || Q->ne[0] > 128) {
constexpr int cols_per_block = 16;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case< 64, cols_per_block>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case< 80, cols_per_block>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case< 96, cols_per_block>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case<112, cols_per_block>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case<128, cols_per_block>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, float>(ctx, dst);
ggml_cuda_flash_attn_ext_mma_f16_case<256, cols_per_block>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
} else {
constexpr int cols_per_block = 32;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
break;
// case 256:
// ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
// break;
default:
GGML_ABORT("fatal error");
break;
}
}
static void ggml_cuda_flash_attn_ext_mma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
const ggml_tensor * Q = dst->src[0];
if (Q->ne[1] <= 8) {
ggml_cuda_flash_attn_ext_mma_f16_switch_hs<8>(ctx, dst);
return;
}
if (Q->ne[1] <= 8 && Q->ne[0] % WARP_SIZE == 0) {
constexpr int cols_per_block = 8;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
if (Q->ne[1] <= 16) {
ggml_cuda_flash_attn_ext_mma_f16_switch_hs<16>(ctx, dst);
return;
}
if (Q->ne[1] <= 32) {
constexpr int cols_per_block = 16;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
ggml_cuda_flash_attn_ext_mma_f16_switch_hs<32>(ctx, dst);
return;
}
constexpr int cols_per_block = 32;
switch (Q->ne[0]) {
case 64:
ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
break;
case 80:
ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
break;
case 96:
ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
break;
case 112:
ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
break;
case 128:
ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
break;
case 256:
ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
break;
default:
GGML_ABORT("fatal error");
break;
}
ggml_cuda_flash_attn_ext_mma_f16_switch_hs<64>(ctx, dst);
}
#define FATTN_VEC_F16_CASE(D, type_K, type_V) \
if (Q->ne[0] == (D) && K->type == (type_K) && V->type == (type_V)) { \
ggml_cuda_flash_attn_ext_vec_f16_case<D, type_K, type_V>(ctx, dst); \
@ -323,11 +236,19 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
}
if (!fp16_mma_available(cc)) {
if (prec == GGML_PREC_DEFAULT) {
if (Q->ne[1] <= 8) {
ggml_cuda_flash_attn_ext_vec_f16(ctx, dst);
} else {
ggml_cuda_flash_attn_ext_tile_f16(ctx, dst);
}
} else {
if (Q->ne[1] <= 8) {
ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
} else {
ggml_cuda_flash_attn_ext_tile_f32(ctx, dst);
}
}
return;
}
@ -341,5 +262,11 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
}
}
// The MMA implementation needs Turing or newer, use the old WMMA code for Volta:
if (cc == GGML_CUDA_CC_VOLTA) {
ggml_cuda_flash_attn_ext_wmma_f16(ctx, dst);
return;
}
ggml_cuda_flash_attn_ext_mma_f16(ctx, dst);
}

33
ggml/src/ggml-cuda/ggml-cuda.cu
View file

@ -40,6 +40,7 @@ bool g_mul_mat_q = true;
#include "ggml-cuda/upscale.cuh"
#include "ggml-cuda/wkv6.cuh"
#include "ggml-cuda/gla.cuh"
#include "ggml.h"
#include <algorithm>
#include <array>
@ -1206,7 +1207,7 @@ static void ggml_cuda_op_mul_mat_cublas(
CUBLAS_CHECK(cublasSetStream(ctx.cublas_handle(id), stream));
if (compute_capability == GGML_CUDA_CC_CDNA) {
if (GGML_CUDA_CC_IS_CDNA(compute_capability)) {
const float alpha = 1.0f;
const float beta = 0.0f;
CUBLAS_CHECK(
@ -1366,8 +1367,6 @@ static void ggml_cuda_op_mul_mat(
const int64_t ne13 = src1->ne[3];
const int64_t nrows1 = ggml_nrows(src1);
GGML_ASSERT(ne03 == ne13);
const int64_t ne0 = dst->ne[0];
const int64_t ne1 = dst->ne[1];
@ -1381,9 +1380,11 @@ static void ggml_cuda_op_mul_mat(
GGML_ASSERT(src1->type == GGML_TYPE_F32 || (src1->ne[2] == 1 && src1->ne[3] == 1));
GGML_ASSERT(ne12 >= ne02 && ne12 % ne02 == 0);
GGML_ASSERT(ne12 % ne02 == 0);
GGML_ASSERT(ne13 % ne03 == 0);
const int64_t i02_divisor = ne12 / ne02;
const int64_t i03_divisor = ne13 / ne03;
const size_t src0_ts = ggml_type_size(src0->type);
const size_t src0_bs = ggml_blck_size(src0->type);
@ -1399,6 +1400,7 @@ static void ggml_cuda_op_mul_mat(
GGML_ASSERT(!(split && ne02 > 1));
GGML_ASSERT(!(split && ne03 > 1));
GGML_ASSERT(!(split && ne02 < ne12));
GGML_ASSERT(!(split && ne03 < ne13));
ggml_tensor_extra_gpu * src0_extra = split ? (ggml_tensor_extra_gpu *) src0->extra : nullptr;
@ -1562,7 +1564,8 @@ static void ggml_cuda_op_mul_mat(
}
// for split tensors the data begins at i0 == i0_offset_low
char * src0_dd_i = dev[id].src0_dd + (i0/i02_divisor) * (ne01*ne00*src0_ts)/src0_bs;
const size_t nbytes_src0_matrix = ne01*ne00*src0_ts / src0_bs;
char * src0_dd_i = dev[id].src0_dd + ((i03/i03_divisor)*ne02 + (i02/i02_divisor)) * nbytes_src0_matrix;
float * src1_ddf_i = dev[id].src1_ddf + (i0*ne11 + src1_col_0) * ne10;
char * src1_ddq_i = dev[id].src1_ddq + src1_ddq_i_offset;
float * dst_dd_i = dev[id].dst_dd + (i0*ne1 + src1_col_0) * (dst_on_device ? ne0 : row_diff);
@ -1606,8 +1609,9 @@ static void ggml_cuda_op_mul_mat(
CUDA_CHECK(cudaGetLastError());
}
if (src1_col_0 == 0 && !src0_is_contiguous && i02 % i02_divisor == 0) {
CUDA_CHECK(ggml_cuda_cpy_tensor_2d(src0_dd_i, src0, i03, i02/i02_divisor, dev[id].row_low, dev[id].row_high, stream));
if (src1_col_0 == 0 && !src0_is_contiguous && i03 % i03_divisor == 0 && i02 % i02_divisor == 0) {
CUDA_CHECK(ggml_cuda_cpy_tensor_2d(
src0_dd_i, src0, i03/i03_divisor, i02/i02_divisor, dev[id].row_low, dev[id].row_high, stream));
}
// do the computation
@ -1751,7 +1755,7 @@ static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, co
beta = &beta_f32;
}
if (ggml_cuda_info().devices[ctx.device].cc == GGML_CUDA_CC_CDNA) {
if (GGML_CUDA_CC_IS_CDNA(ggml_cuda_info().devices[ctx.device].cc)) {
cu_compute_type = CUBLAS_COMPUTE_32F;
alpha = &alpha_f32;
beta = &beta_f32;
@ -1882,7 +1886,7 @@ static void ggml_cuda_mul_mat(ggml_backend_cuda_context & ctx, const ggml_tensor
//printf("src0 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src0), ggml_is_transposed(src0), ggml_type_name(src0->type), src0->name);
//printf("src1 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src1), ggml_is_transposed(src1), ggml_type_name(src1->type), src1->name);
if (!split && use_mul_mat_vec && dst->ne[3] == 1 && (src0->ne[1] < MMV_MAX_ROWS || any_gpus_without_fp16_mma)) {
if (!split && use_mul_mat_vec && (src0->ne[1] < MMV_MAX_ROWS || any_gpus_without_fp16_mma)) {
// the custom F16 vector kernel can be used over batched cuBLAS GEMM
// but this is only faster for GPUs without tensor cores or with a thin src0 matrix (particularly KQV in attention)
ggml_cuda_mul_mat_vec(ctx, src0, src1, dst);
@ -2220,12 +2224,7 @@ static bool ggml_cuda_compute_forward(ggml_backend_cuda_context & ctx, struct gg
ggml_cuda_op_rms_norm_back(ctx, dst);
break;
case GGML_OP_MUL_MAT:
if (dst->src[0]->ne[3] != dst->src[1]->ne[3]) {
GGML_LOG_ERROR("%s: cannot compute %s: src0->ne[3] = %" PRId64 ", src1->ne[3] = %" PRId64 " - fallback to CPU\n", __func__, dst->name, dst->src[0]->ne[3], dst->src[1]->ne[3]);
return false;
} else {
ggml_cuda_mul_mat(ctx, dst->src[0], dst->src[1], dst);
}
break;
case GGML_OP_MUL_MAT_ID:
ggml_cuda_mul_mat_id(ctx, dst);
@ -3002,9 +3001,6 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
if (b->type == GGML_TYPE_F16 && a->type != GGML_TYPE_F16) {
return false;
}
if (op->op == GGML_OP_MUL_MAT && a->ne[3] != b->ne[3]) {
return false;
}
#ifdef GGML_USE_MUSA
if (b->type == GGML_TYPE_F16 && b->ne[2]*b->ne[3] > 1 &&
!ggml_is_transposed(a) && !ggml_is_transposed(b)) {
@ -3144,6 +3140,7 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
break;
case GGML_OP_NORM:
case GGML_OP_RMS_NORM:
return true;
case GGML_OP_RMS_NORM_BACK:
return ggml_is_contiguous(op->src[0]) && op->ne[0] % WARP_SIZE == 0;
break;
@ -3186,7 +3183,9 @@ static bool ggml_backend_cuda_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_SUM_ROWS:
case GGML_OP_ARGSORT:
case GGML_OP_ACC:
return true;
case GGML_OP_GROUP_NORM:
return ggml_is_contiguous(op->src[0]);
case GGML_OP_UPSCALE:
case GGML_OP_PAD:
case GGML_OP_ARANGE:

335
ggml/src/ggml-cuda/mma.cuh
View file

@ -1,11 +1,67 @@
// This file contains primitives that expose the tensor core PTX instructions for CUDA code.
// The primitives can be used in a similar way as the nvcuda::wmma interface but with a well-defined memory layout.
// The documentation for the PTX instructions can be found under:
// https://docs.nvidia.com/cuda/parallel-thread-execution/index.html#matrix-multiply-accumulate-operation-using-mma-instruction
//
// Like with nvcuda::wmma there are three types of matrix tiles: A, B, and C with A @ B = C.
// A is a row-major matrix with shape I x K.
// B is a column-major matrix with shape K x J.
// C is a column-major matrix with shape I x J.
// Note that along their lowest dimension I, J, and K are measured in physical 32 bit elements instead of logical elements.
// The functions get_i, get_j, and get_k can be used to get the physical 32 bit index of the lth element of a thread within a tile.
// All matrix tiles have ne physical 32 bit elements per warp.
//
// As described in the documentation, all pointers for load_ldmatrix must be to shared memory and aligned to 16 bytes.
#include "common.cuh"
struct mma_int_A_I16K4 {
#if CUDART_VERSION >= 11800
static __device__ __forceinline__ int ggml_cuda_movmatrix(const int x) {
int ret = 0;
#ifdef NEW_MMA_AVAILABLE
asm("movmatrix.sync.aligned.m8n8.trans.b16 %0, %1;"
: "+r"(ret) : "r"(x));
#else
NO_DEVICE_CODE;
#endif // defined(NEW_MMA_AVAILABLE)
return ret;
}
#else
static __device__ __forceinline__ int ggml_cuda_movmatrix(const int x) {
// Imagine transposing row-major matrix to column-major matrix.
const int src_i_low = 2 * (threadIdx.x % 4);
const int src_i_high = src_i_low + 1;
const int src_j = threadIdx.x / 4;
const int src_laneid_low = src_i_low * 4 + src_j / 2;
const int src_laneid_high = src_i_high * 4 + src_j / 2;
const int shift_low = ((src_j + 0) % 2) * 16;
const int shift_high = ((src_j + 1) % 2) * 16;
const int ret_low = (__shfl_sync(0xFFFFFFFF, x, src_laneid_low, WARP_SIZE) >> shift_low) & 0x0000FFFF;
const int ret_high = (__shfl_sync(0xFFFFFFFF, x, src_laneid_high, WARP_SIZE) << shift_high) & 0xFFFF0000;
return ret_low | ret_high;
}
#endif // CUDART_VERSION >= 11800
template <typename T>
struct mma_A_I16K4 {
static_assert(sizeof(T) == 4, "bad type size");
static constexpr int I = 16;
static constexpr int K = 4;
static constexpr int ne = 2;
int x[ne] = {0};
T x[ne];
static __device__ __forceinline__ int get_i(const int l) {
const int ret = (l%2) * (I/2) + threadIdx.x / K;
@ -21,27 +77,35 @@ struct mma_int_A_I16K4 {
return ret;
}
__device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE)
const int * xs = xs0 + (threadIdx.x%I)*stride;
asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
: "+r"(x[0]), "+r"(x[1])
: "l"(xs));
#else
__device__ __forceinline__ void load_generic(const T * __restrict__ xs0, const int & stride) {
#pragma unroll
for (int l = 0; l < ne; ++l) {
x[l] = xs0[get_i(l)*stride + get_k(l)];
}
#endif // defined(INT8_MMA_AVAILABLE)
}
__device__ __forceinline__ void load_ldmatrix(const T * __restrict__ xs0, const int & stride) {
#ifdef NEW_MMA_AVAILABLE
int * xi = (int *) x;
const int * xs = (const int *) xs0 + (threadIdx.x%I)*stride;
asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
: "+r"(xi[0]), "+r"(xi[1])
: "l"(xs));
#else
load_generic(xs0, stride);
#endif // NEW_MMA_AVAILABLE
}
};
struct mma_int_A_I16K8 {
template <typename T>
struct mma_A_I16K8 {
static_assert(sizeof(T) == 4, "bad type size");
static constexpr int I = 16;
static constexpr int K = 8;
static constexpr int ne = 4;
int x[ne] = {0};
T x[ne];
static __device__ __forceinline__ int get_i(const int l) {
const int ret = (l%2) * (I/2) + threadIdx.x / (K/2);
@ -57,31 +121,62 @@ struct mma_int_A_I16K8 {
return ret;
}
__device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE)
const int * xs = xs0 + (threadIdx.x%I)*stride + (threadIdx.x/I)*(K/2);
asm("ldmatrix.sync.aligned.m8n8.x4.b16 {%0, %1, %2, %3}, [%4];"
: "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
: "l"(xs));
#else
__device__ __forceinline__ void load_generic(const T * __restrict__ xs0, const int & stride) {
#pragma unroll
for (int l = 0; l < ne; ++l) {
x[l] = xs0[get_i(l)*stride + get_k(l)];
}
#endif // defined(INT8_MMA_AVAILABLE)
}
__device__ __forceinline__ void load_low(const int * __restrict__ xs0, const int & stride) {
((mma_int_A_I16K4 *) x)[0].load(xs0, stride);
__device__ __forceinline__ void load_ldmatrix(const T * __restrict__ xs0, const int & stride) {
#ifdef NEW_MMA_AVAILABLE
int * xi = (int * ) x;
const int * xs = (const int *) xs0 + (threadIdx.x%I)*stride + (threadIdx.x/I)*(K/2);
asm("ldmatrix.sync.aligned.m8n8.x4.b16 {%0, %1, %2, %3}, [%4];"
: "+r"(xi[0]), "+r"(xi[1]), "+r"(xi[2]), "+r"(xi[3])
: "l"(xs));
#else
GGML_UNUSED(xs0);
GGML_UNUSED(stride);
NO_DEVICE_CODE;
#endif // NEW_MMA_AVAILABLE
}
__device__ __forceinline__ void load_ldmatrix_trans(const T * __restrict__ xs0, const int & stride) {
#ifdef NEW_MMA_AVAILABLE
int * xi = (int * ) x;
const int * xs = (const int *) xs0 + (threadIdx.x%I)*stride + (threadIdx.x/I)*(K/2);
asm("ldmatrix.sync.aligned.m8n8.x4.trans.b16 {%0, %1, %2, %3}, [%4];"
: "+r"(xi[0]), "+r"(xi[2]), "+r"(xi[1]), "+r"(xi[3])
: "l"(xs));
#else
GGML_UNUSED(xs0);
GGML_UNUSED(stride);
NO_DEVICE_CODE;
#endif // NEW_MMA_AVAILABLE
}
__device__ __forceinline__ void transpose() {
int * xi = (int *) x;
xi[0] = ggml_cuda_movmatrix(xi[0]);
const int tmp = ggml_cuda_movmatrix(xi[1]);
xi[1] = ggml_cuda_movmatrix(xi[2]);
xi[2] = tmp;
xi[3] = ggml_cuda_movmatrix(xi[3]);
}
};
struct mma_int_B_J8K4 {
template <typename T>
struct mma_B_J8K4 {
static_assert(sizeof(T) == 4, "bad type size");
static constexpr int J = 8;
static constexpr int K = 4;
static constexpr int ne = 1;
int x[ne] = {0};
T x[ne];
static __device__ __forceinline__ int get_j(const int /* l */) {
const int ret = threadIdx.x / K;
@ -97,27 +192,34 @@ struct mma_int_B_J8K4 {
return ret;
}
__device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE) && false // Loading as 4 byte values is faster
const int * xs = xs0 + (threadIdx.x%J)*stride;
asm("ldmatrix.sync.aligned.m8n8.x1.b16 {%0}, [%1];"
: "+r"(x[0])
: "l"(xs));
#else
__device__ __forceinline__ void load_generic(const T * __restrict__ xs0, const int & stride) {
#pragma unroll
for (int l = 0; l < ne; ++l) {
x[l] = xs0[get_j(l)*stride + get_k(l)];
}
#endif // defined(INT8_MMA_AVAILABLE)
}
__device__ __forceinline__ void load_ldmatrix(const T * __restrict__ xs0, const int & stride) {
#ifdef NEW_MMA_AVAILABLE
int * xi = (int *) x;
const int * xs = (const int *) xs0 + (threadIdx.x%J)*stride;
asm("ldmatrix.sync.aligned.m8n8.x1.b16 {%0}, [%1];"
: "+r"(xi[0]) : "l"(xs));
#else
load_generic(xs0, stride);
#endif // NEW_MMA_AVAILABLE
}
};
struct mma_int_B_J8K8 {
template <typename T>
struct mma_B_J8K8 {
static_assert(sizeof(T) == 4, "bad type size");
static constexpr int J = 8;
static constexpr int K = 8;
static constexpr int ne = 2;
int x[ne] = {0};
T x[ne];
static __device__ __forceinline__ int get_j(const int /* l */) {
const int ret = threadIdx.x / (K/2);
@ -133,22 +235,31 @@ struct mma_int_B_J8K8 {
return ret;
}
__device__ __forceinline__ void load(const int * __restrict__ xs0, const int & stride) {
#if defined(INT8_MMA_AVAILABLE) && false // Loading as 4 byte values is faster
const int * xs = xs0 + (threadIdx.x%J)*stride + ((threadIdx.x/J)*(K/2)) % K;
asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
: "+r"(x[0]), "+r"(x[1])
: "l"(xs));
#else
__device__ __forceinline__ void load_generic(const T * __restrict__ xs0, const int & stride) {
#pragma unroll
for (int l = 0; l < ne; ++l) {
x[l] = xs0[get_j(l)*stride + get_k(l)];
}
#endif // defined(INT8_MMA_AVAILABLE)
}
__device__ __forceinline__ void load_ldmatrix(const T * __restrict__ xs0, const int & stride) {
#ifdef NEW_MMA_AVAILABLE
int * xi = (int *) x;
const int * xs = (const int *) xs0 + (threadIdx.x%J)*stride + ((threadIdx.x/J)*(K/2)) % K;
asm("ldmatrix.sync.aligned.m8n8.x2.b16 {%0, %1}, [%2];"
: "+r"(xi[0]), "+r"(xi[1])
: "l"(xs));
#else
load_generic(xs0, stride);
#endif // NEW_MMA_AVAILABLE
}
};
struct mma_int_C_I16J8 {
template <typename T>
struct mma_C_I16J8 {};
template <>
struct mma_C_I16J8<int> {
static constexpr int I = 16;
static constexpr int J = 8;
static constexpr int ne = 4;
@ -169,8 +280,8 @@ struct mma_int_C_I16J8 {
return ret;
}
__device__ __forceinline__ void mma_K4(const mma_int_A_I16K4 & mma_A, const mma_int_B_J8K4 & mma_B) {
#ifdef INT8_MMA_AVAILABLE
__device__ __forceinline__ void mma(const mma_A_I16K4<int> & mma_A, const mma_B_J8K4<int> & mma_B) {
#ifdef NEW_MMA_AVAILABLE
#if __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
asm("mma.sync.aligned.m16n8k16.row.col.s32.s8.s8.s32 {%0, %1, %2, %3}, {%4, %5}, {%6}, {%0, %1, %2, %3};"
: "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
@ -188,11 +299,11 @@ struct mma_int_C_I16J8 {
GGML_UNUSED(mma_A);
GGML_UNUSED(mma_B);
NO_DEVICE_CODE;
#endif // INT8_MMA_AVAILABLE
#endif // NEW_MMA_AVAILABLE
}
__device__ __forceinline__ void mma_K8(const mma_int_A_I16K8 & mma_A, const mma_int_B_J8K8 & mma_B) {
#ifdef INT8_MMA_AVAILABLE
__device__ __forceinline__ void mma(const mma_A_I16K8<int> & mma_A, const mma_B_J8K8<int> & mma_B) {
#ifdef NEW_MMA_AVAILABLE
#if __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
asm("mma.sync.aligned.m16n8k32.row.col.s32.s8.s8.s32 {%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3};"
: "+r"(x[0]), "+r"(x[1]), "+r"(x[2]), "+r"(x[3])
@ -216,6 +327,132 @@ struct mma_int_C_I16J8 {
GGML_UNUSED(mma_A);
GGML_UNUSED(mma_B);
NO_DEVICE_CODE;
#endif // INT8_MMA_AVAILABLE
#endif // NEW_MMA_AVAILABLE
}
};
template <>
struct mma_C_I16J8<half2> {
static constexpr int I = 16;
static constexpr int J = 4;
static constexpr int ne = 2;
half2 x[ne] = {{0.0f, 0.0f}, {0.0f, 0.0f}};
static __device__ __forceinline__ int get_i(const int l) {
const int ret = l * (I/2) + threadIdx.x / J;
GGML_CUDA_ASSUME(ret >= 0);
GGML_CUDA_ASSUME(ret < I);
return ret;
}
static __device__ __forceinline__ int get_j(const int /* l */) {
const int ret = threadIdx.x % J;
GGML_CUDA_ASSUME(ret >= 0);
GGML_CUDA_ASSUME(ret < J);
return ret;
}
__device__ __forceinline__ void mma(const mma_A_I16K8<half2> & mma_A, const mma_B_J8K8<half2> & mma_B) {
#ifdef NEW_MMA_AVAILABLE
int * Axi = (int *) mma_A.x;
int * Bxi = (int *) mma_B.x;
int * xi = (int *) x;
#if __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
asm("mma.sync.aligned.m16n8k16.row.col.f16.f16.f16.f16 {%0, %1}, {%2, %3, %4, %5}, {%6, %7}, {%0, %1};"
: "+r"(xi[0]), "+r"(xi[1])
: "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]));
#else
// On Turing m16n8k16 mma is not available, use 2x m8n8k8 mma instead:
asm("mma.sync.aligned.m16n8k8.row.col.f16.f16.f16.f16 {%0, %1}, {%2, %3}, {%4}, {%0, %1};"
: "+r"(xi[0]), "+r"(xi[1])
: "r"(Axi[0]), "r"(Axi[1]), "r"(Bxi[0]));
asm("mma.sync.aligned.m16n8k8.row.col.f16.f16.f16.f16 {%0, %1}, {%2, %3}, {%4}, {%0, %1};"
: "+r"(xi[0]), "+r"(xi[1])
: "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[1]));
#endif // __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
#else
GGML_UNUSED(mma_A);
GGML_UNUSED(mma_B);
NO_DEVICE_CODE;
#endif // NEW_MMA_AVAILABLE
}
__device__ __forceinline__ mma_B_J8K8<half2> to_mma_B() {
mma_B_J8K8<half2> mma_B;
int * xi = (int *) x;
int * Bxi = (int *) mma_B.x;
Bxi[0] = ggml_cuda_movmatrix(xi[0]);
Bxi[1] = ggml_cuda_movmatrix(xi[1]);
return mma_B;
}
};
template <>
struct mma_C_I16J8<float> {
static constexpr int I = 16;
static constexpr int J = 8;
static constexpr int ne = 4;
float x[ne] = {0.0f, 0.0f, 0.0f, 0.0f};
static __device__ __forceinline__ int get_i(const int l) {
const int ret = (l/2) * (I/2) + threadIdx.x / (J/2);
GGML_CUDA_ASSUME(ret >= 0);
GGML_CUDA_ASSUME(ret < I);
return ret;
}
static __device__ __forceinline__ int get_j(const int l) {
const int ret = 2 * (threadIdx.x % (J/2)) + l%2;
GGML_CUDA_ASSUME(ret >= 0);
GGML_CUDA_ASSUME(ret < J);
return ret;
}
__device__ __forceinline__ void mma(const mma_A_I16K8<half2> & mma_A, const mma_B_J8K8<half2> & mma_B) {
#ifdef NEW_MMA_AVAILABLE
int * Axi = (int *) mma_A.x;
int * Bxi = (int *) mma_B.x;
int * xi = (int *) x;
#if __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
asm("mma.sync.aligned.m16n8k16.row.col.f32.f16.f16.f32 {%0, %1, %2, %3}, {%4, %5, %6, %7}, {%8, %9}, {%0, %1, %2, %3};"
: "+r"(xi[0]), "+r"(xi[1]), "+r"(xi[2]), "+r"(xi[3])
: "r"(Axi[0]), "r"(Axi[1]), "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[0]), "r"(Bxi[1]));
#else
// On Turing m16n8k16 mma is not available, use 2x m8n8k8 mma instead:
asm("mma.sync.aligned.m16n8k8.row.col.f32.f16.f16.f32 {%0, %1, %2, %3}, {%4, %5}, {%6}, {%0, %1, %2, %3};"
: "+r"(xi[0]), "+r"(xi[1]), "+r"(xi[2]), "+r"(xi[3])
: "r"(Axi[0]), "r"(Axi[1]), "r"(Bxi[0]));
asm("mma.sync.aligned.m16n8k8.row.col.f32.f16.f16.f32 {%0, %1, %2, %3}, {%4, %5}, {%6}, {%0, %1, %2, %3};"
: "+r"(xi[0]), "+r"(xi[1]), "+r"(xi[2]), "+r"(xi[3])
: "r"(Axi[2]), "r"(Axi[3]), "r"(Bxi[1]));
#endif // __CUDA_ARCH__ >= GGML_CUDA_CC_AMPERE
#else
GGML_UNUSED(mma_A);
GGML_UNUSED(mma_B);
NO_DEVICE_CODE;
#endif // NEW_MMA_AVAILABLE
}
__device__ __forceinline__ mma_B_J8K8<half2> to_mma_B() {
mma_B_J8K8<half2> mma_B;
mma_B.x[0] = make_half2(x[0], x[1]);
mma_B.x[1] = make_half2(x[2], x[3]);
int * Bxi = (int *) mma_B.x;
Bxi[0] = ggml_cuda_movmatrix(Bxi[0]);
Bxi[1] = ggml_cuda_movmatrix(Bxi[1]);
return mma_B;
}
__device__ __forceinline__ void load_generic(const float * __restrict__ xs0, const int & stride) {
#pragma unroll
for (int l = 0; l < ne; ++l) {
x[l] = xs0[get_j(l)*stride + get_i(l)];
}
}
};

4
ggml/src/ggml-cuda/mmq.cu
View file

@ -134,7 +134,7 @@ bool ggml_cuda_should_use_mmq(enum ggml_type type, int cc, int64_t ne11) {
return false;
}
if (int8_mma_available(cc)) {
if (new_mma_available(cc)) {
return true;
}
@ -150,5 +150,5 @@ bool ggml_cuda_should_use_mmq(enum ggml_type type, int cc, int64_t ne11) {
return cc < GGML_CUDA_CC_VOLTA || ne11 < MMQ_DP4A_MAX_BATCH_SIZE;
}
return (cc < GGML_CUDA_CC_RDNA3 && cc != GGML_CUDA_CC_CDNA && cc != GGML_CUDA_CC_VEGA20) || ne11 < MMQ_DP4A_MAX_BATCH_SIZE;
return (!GGML_CUDA_CC_IS_RDNA3(cc) && !GGML_CUDA_CC_IS_CDNA(cc) && !GGML_CUDA_CC_IS_GCN(cc)) || ne11 < MMQ_DP4A_MAX_BATCH_SIZE;
}

351
ggml/src/ggml-cuda/mmq.cuh
View file

File diff suppressed because it is too large Load diff

146
ggml/src/ggml-cuda/mmv.cu
View file

@ -1,25 +1,29 @@
#include "ggml.h"
#include "common.cuh"
#include "mmv.cuh"
template <typename T, typename type_acc, int block_size>
static __global__ void mul_mat_vec(
const T * __restrict__ x, const float * __restrict__ y, float * __restrict__ dst, const int64_t ncols2, const int64_t stride_row,
const int64_t channel_ratio, const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst) {
const int64_t channel_ratio, const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst,
const int64_t sample_ratio, const int64_t stride_sample_x, const int64_t stride_sample_y, const int64_t stride_sample_dst) {
const int64_t row = blockIdx.x;
const int64_t channel = blockIdx.z;
const int64_t channel = blockIdx.y;
const int64_t sample = blockIdx.z;
const int tid = threadIdx.x;
constexpr int warp_size = ggml_cuda_get_physical_warp_size();
x += (channel/channel_ratio)*stride_channel_x + row*stride_row;
y += channel *stride_channel_y;
dst += channel *stride_channel_dst;
x += (sample/sample_ratio)*stride_sample_x + (channel/channel_ratio)*stride_channel_x + row*stride_row;
y += sample *stride_sample_y + channel *stride_channel_y;
dst += sample *stride_sample_dst + channel *stride_channel_dst;
const float2 * y2 = (const float2 *) y;
extern __shared__ char data_mmv[];
float * buf_iw = (float *) data_mmv;
if (block_size > WARP_SIZE) {
if (tid < WARP_SIZE) {
if (block_size > warp_size) {
if (tid < warp_size) {
buf_iw[tid] = 0.0f;
}
__syncthreads();
@ -67,16 +71,16 @@ static __global__ void mul_mat_vec(
static_assert(std::is_same<T, void>::value, "unsupported type");
}
sumf = warp_reduce_sum(sumf);
sumf = warp_reduce_sum<warp_size>(sumf);
if (block_size > WARP_SIZE) {
buf_iw[tid/WARP_SIZE] = sumf;
if (block_size > warp_size) {
buf_iw[tid/warp_size] = sumf;
__syncthreads();
if (tid >= WARP_SIZE) {
if (tid >= warp_size) {
return;
}
sumf = buf_iw[tid];
sumf = warp_reduce_sum(sumf);
sumf = warp_reduce_sum<warp_size>(sumf);
}
if (tid != 0) {
@ -90,16 +94,28 @@ template <typename T, typename type_acc>
static void launch_mul_mat_vec_cuda(
const T * x, const float * y, float * dst,
const int64_t ncols, const int64_t nrows, const int64_t stride_row, const int64_t nchannels_x, const int64_t nchannels_y,
const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst,
const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst, const int64_t nsamples_x,
const int64_t nsamples_y, const int64_t stride_sample_x, const int64_t stride_sample_y, const int64_t stride_sample_dst,
cudaStream_t stream) {
GGML_ASSERT(ncols % 2 == 0);
GGML_ASSERT(stride_row % 2 == 0);
GGML_ASSERT(nchannels_y % nchannels_x == 0);
GGML_ASSERT(nsamples_y % nsamples_x == 0);
const int64_t channel_ratio = nchannels_y / nchannels_x;
const int64_t sample_ratio = nsamples_y / nsamples_x;
int device;
int warp_size;
int64_t block_size_best = WARP_SIZE;
int64_t niter_best = (ncols + 2*WARP_SIZE - 1) / (2*WARP_SIZE);
for (int64_t block_size = 2*WARP_SIZE; block_size <= 256; block_size += WARP_SIZE) {
CUDA_CHECK(cudaGetDevice(&device));
warp_size = ggml_cuda_info().devices[device].warp_size;
int64_t block_size_best = warp_size;
int64_t niter_best = (ncols + 2*warp_size - 1) / (2*warp_size);
int64_t max_block_size = 256;
if(ggml_cuda_info().devices[device].cc > GGML_CUDA_CC_OFFSET_AMD && ggml_cuda_info().devices[device].cc < GGML_CUDA_CC_RDNA1) {
max_block_size = 128;
}
for (int64_t block_size = 2*warp_size; block_size <= max_block_size; block_size += warp_size) {
const int64_t niter = (ncols + 2*block_size - 1) / (2*block_size);
if (niter < niter_best) {
niter_best = niter;
@ -107,41 +123,49 @@ static void launch_mul_mat_vec_cuda(
}
}
const int smem = WARP_SIZE*sizeof(float);
const dim3 block_nums(nrows, 1, nchannels_y);
const int smem = warp_size*sizeof(float);
const dim3 block_nums(nrows, nchannels_y, nsamples_y);
const dim3 block_dims(block_size_best, 1, 1);
switch (block_size_best) {
case 32: {
mul_mat_vec<T, type_acc, 32><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 64: {
mul_mat_vec<T, type_acc, 64><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 96: {
mul_mat_vec<T, type_acc, 96><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 128: {
mul_mat_vec<T, type_acc, 128><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 160: {
mul_mat_vec<T, type_acc, 160><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 192: {
mul_mat_vec<T, type_acc, 192><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 224: {
mul_mat_vec<T, type_acc, 224><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
case 256: {
mul_mat_vec<T, type_acc, 256><<<block_nums, block_dims, smem, stream>>>
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst);
(x, y, dst, ncols/2, stride_row, channel_ratio, stride_channel_x, stride_channel_y, stride_channel_dst,
sample_ratio, stride_sample_x, stride_sample_y, stride_sample_dst);
} break;
default: {
GGML_ABORT("fatal error");
@ -153,16 +177,19 @@ template<typename T>
static void mul_mat_vec_cuda(
const T * x, const float * y, float * dst,
const int64_t ncols, const int64_t nrows, const int64_t stride_row, const int64_t nchannels_x, const int64_t nchannels_y,
const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst,
const int64_t stride_channel_x, const int64_t stride_channel_y, const int64_t stride_channel_dst, const int64_t nsamples_x,
const int64_t nsamples_y, const int64_t stride_sample_x, const int64_t stride_sample_y, const int64_t stride_sample_dst,
enum ggml_prec prec, cudaStream_t stream) {
switch (prec) {
case GGML_PREC_DEFAULT: {
launch_mul_mat_vec_cuda<T, half>(x, y, dst, ncols, nrows, stride_row, nchannels_x, nchannels_y,
stride_channel_x, stride_channel_y, stride_channel_dst, stream);
launch_mul_mat_vec_cuda<T, half>
(x, y, dst, ncols, nrows, stride_row, nchannels_x, nchannels_y, stride_channel_x, stride_channel_y, stride_channel_dst,
nsamples_x, nsamples_y, stride_sample_x, stride_sample_y, stride_sample_dst, stream);
} break;
case GGML_PREC_F32: {
launch_mul_mat_vec_cuda<T, float>(x, y, dst, ncols, nrows, stride_row, nchannels_x, nchannels_y,
stride_channel_x, stride_channel_y, stride_channel_dst, stream);
launch_mul_mat_vec_cuda<T, float>
(x, y, dst, ncols, nrows, stride_row, nchannels_x, nchannels_y, stride_channel_x, stride_channel_y, stride_channel_dst,
nsamples_x, nsamples_y, stride_sample_x, stride_sample_y, stride_sample_dst, stream);
} break;
}
}
@ -171,10 +198,19 @@ void ggml_cuda_mul_mat_vec(ggml_backend_cuda_context & ctx, const ggml_tensor *
GGML_ASSERT(src1->type == GGML_TYPE_F32);
GGML_ASSERT(dst->type == GGML_TYPE_F32);
const int64_t ne00 = src0->ne[0];
const int64_t ne01 = src0->ne[1];
GGML_TENSOR_BINARY_OP_LOCALS;
GGML_ASSERT(src1->ne[1] == 1);
const size_t ts_src0 = ggml_type_size(src0->type);
const size_t ts_src1 = ggml_type_size(src1->type);
const size_t ts_dst = ggml_type_size(dst->type);
GGML_ASSERT(ne11 == 1);
GGML_ASSERT(ne12 == ne2);
GGML_ASSERT(ne13 == ne3);
GGML_ASSERT(nb00 == ts_src0);
GGML_ASSERT(nb10 == ts_src1);
GGML_ASSERT(nb0 == ts_dst);
const int cc = ggml_cuda_info().devices[ggml_cuda_get_device()].cc;
const enum ggml_prec prec = fast_fp16_available(cc) ? ggml_prec(dst->op_params[0]) : GGML_PREC_F32;
@ -182,29 +218,22 @@ void ggml_cuda_mul_mat_vec(ggml_backend_cuda_context & ctx, const ggml_tensor *
const float * src1_d = (const float *) src1->data;
float * dst_d = (float *) dst->data;
const int64_t ne02 = src0->ne[2];
const int64_t ne12 = src1->ne[2];
GGML_ASSERT(dst->ne[2] == ne12);
GGML_ASSERT(src0->ne[3] == 1);
GGML_ASSERT(src1->ne[3] == 1);
GGML_ASSERT( dst->ne[3] == 1);
const int64_t stride_row = src0->nb[1] / ggml_type_size(src0->type);
const int64_t channel_stride_x = src0->nb[2] / ggml_type_size(src0->type);
const int64_t channel_stride_y = src1->nb[2] / ggml_type_size(src1->type);
const int64_t channel_stride_dst = dst->nb[2] / ggml_type_size( dst->type);
const int64_t s01 = src0->nb[1] / ts_src0;
const int64_t s02 = src0->nb[2] / ts_src0;
const int64_t s12 = src1->nb[2] / ts_src1;
const int64_t s2 = dst->nb[2] / ts_dst;
const int64_t s03 = src0->nb[3] / ts_src0;
const int64_t s13 = src1->nb[3] / ts_src1;
const int64_t s3 = dst->nb[3] / ts_dst;
switch (src0->type) {
case GGML_TYPE_F16: {
const half * src0_d = (const half *) src0->data;
mul_mat_vec_cuda(src0_d, src1_d, dst_d, ne00, ne01, stride_row, ne02, ne12,
channel_stride_x, channel_stride_y, channel_stride_dst, prec, ctx.stream());
mul_mat_vec_cuda(src0_d, src1_d, dst_d, ne00, ne01, s01, ne02, ne12, s02, s12, s2, ne03, ne13, s03, s13, s3, prec, ctx.stream());
} break;
case GGML_TYPE_BF16: {
const nv_bfloat16 * src0_d = (const nv_bfloat16 *) src0->data;
mul_mat_vec_cuda(src0_d, src1_d, dst_d, ne00, ne01, stride_row, ne02, ne12,
channel_stride_x, channel_stride_y, channel_stride_dst, prec, ctx.stream());
mul_mat_vec_cuda(src0_d, src1_d, dst_d, ne00, ne01, s01, ne02, ne12, s02, s12, s2, ne03, ne13, s03, s13, s3, prec, ctx.stream());
} break;
default:
GGML_ABORT("unsupported type: %s", ggml_type_name(src0->type));
@ -233,20 +262,27 @@ void ggml_cuda_op_mul_mat_vec(
const int64_t stride_row = ne00;
const int64_t nchannels_x = 1;
const int64_t nchannels_y = 1;
const int64_t channel_stride_x = 0;
const int64_t channel_stride_y = 0;
const int64_t channel_stride_dst = 0;
const int64_t stride_channel_x = 0;
const int64_t stride_channel_y = 0;
const int64_t stride_channel_dst = 0;
const int64_t nsamples_x = 1;
const int64_t nsamples_y = 1;
const int64_t stride_sample_x = 0;
const int64_t stride_sample_y = 0;
const int64_t stride_sample_dst = 0;
switch (src0->type) {
case GGML_TYPE_F16: {
const half * src0_d = (const half *) src0_dd_i;
mul_mat_vec_cuda(src0_d, src1_ddf_i, dst_dd_i, ne00, row_diff, stride_row,
nchannels_x, nchannels_y, channel_stride_x, channel_stride_y, channel_stride_dst, prec, stream);
nchannels_x, nchannels_y, stride_channel_x, stride_channel_y, stride_channel_dst,
nsamples_x, nsamples_y, stride_sample_x, stride_sample_y, stride_sample_dst, prec, stream);
} break;
case GGML_TYPE_BF16: {
const nv_bfloat16 * src0_d = (const nv_bfloat16 *) src0_dd_i;
mul_mat_vec_cuda(src0_d, src1_ddf_i, dst_dd_i, ne00, row_diff, stride_row,
nchannels_x, nchannels_y, channel_stride_x, channel_stride_y, channel_stride_dst, prec, stream);
nchannels_x, nchannels_y, stride_channel_x, stride_channel_y, stride_channel_dst,
nsamples_x, nsamples_y, stride_sample_x, stride_sample_y, stride_sample_dst, prec, stream);
} break;
default:
GGML_ABORT("unsupported type: %s", ggml_type_name(src0->type));

77
ggml/src/ggml-cuda/norm.cu
View file

@ -1,12 +1,20 @@
#include "norm.cuh"
#include <cstdint>
template <int block_size>
static __global__ void norm_f32(const float * x, float * dst, const int ncols, const float eps) {
const int row = blockIdx.x*blockDim.y + threadIdx.y;
static __global__ void norm_f32(
const float * x, float * dst, const int ncols, const int64_t stride_row, const int64_t stride_channel,
const int64_t stride_sample, const float eps) {
const int nrows = gridDim.x;
const int nchannels = gridDim.y;
const int row = blockIdx.x;
const int channel = blockIdx.y;
const int sample = blockIdx.z;
const int tid = threadIdx.x;
x += int64_t(row)*ncols;
dst += int64_t(row)*ncols;
x += sample*stride_sample + channel*stride_channel + row*stride_row;
dst += ((sample*nchannels + channel)*nrows + row)*ncols;
float2 mean_var = make_float2(0.0f, 0.0f);
@ -97,12 +105,19 @@ static __global__ void group_norm_f32(const float * x, float * dst, const int gr
}
template <int block_size>
static __global__ void rms_norm_f32(const float * x, float * dst, const int ncols, const float eps) {
const int row = blockIdx.x*blockDim.y + threadIdx.y;
static __global__ void rms_norm_f32(
const float * x, float * dst, const int ncols, const int64_t stride_row, const int64_t stride_channel,
const int64_t stride_sample, const float eps) {
const int nrows = gridDim.x;
const int nchannels = gridDim.y;
const int row = blockIdx.x;
const int channel = blockIdx.y;
const int sample = blockIdx.z;
const int tid = threadIdx.x;
x += int64_t(row)*ncols;
dst += int64_t(row)*ncols;
x += sample*stride_sample + channel*stride_channel + row*stride_row;
dst += ((sample*nchannels + channel)*nrows + row)*ncols;
float tmp = 0.0f; // partial sum for thread in warp
@ -186,13 +201,16 @@ static __global__ void rms_norm_back_f32(
}
}
static void norm_f32_cuda(const float * x, float * dst, const int ncols, const int nrows, const float eps, cudaStream_t stream) {
static void norm_f32_cuda(
const float * x, float * dst, const int ncols, const int nrows, const int nchannels, const int nsamples,
const int64_t stride_row, const int64_t stride_channel, const int64_t stride_sample, const float eps, cudaStream_t stream) {
const dim3 blocks_num(nrows, nchannels, nsamples);
if (ncols < 1024) {
const dim3 block_dims(WARP_SIZE, 1, 1);
norm_f32<WARP_SIZE><<<nrows, block_dims, 0, stream>>>(x, dst, ncols, eps);
norm_f32<WARP_SIZE><<<blocks_num, block_dims, 0, stream>>>(x, dst, ncols, stride_row, stride_channel, stride_sample, eps);
} else {
const dim3 block_dims(1024, 1, 1);
norm_f32<1024><<<nrows, block_dims, 0, stream>>>(x, dst, ncols, eps);
norm_f32<1024><<<blocks_num, block_dims, 0, stream>>>(x, dst, ncols, stride_row, stride_channel, stride_sample, eps);
}
}
@ -207,13 +225,16 @@ static void group_norm_f32_cuda(
}
}
static void rms_norm_f32_cuda(const float * x, float * dst, const int ncols, const int nrows, const float eps, cudaStream_t stream) {
static void rms_norm_f32_cuda(
const float * x, float * dst, const int ncols, const int nrows, const int nchannels, const int nsamples,
const int64_t stride_row, const int64_t stride_channel, const int64_t stride_sample, const float eps, cudaStream_t stream) {
const dim3 blocks_num(nrows, nchannels, nsamples);
if (ncols < 1024) {
const dim3 block_dims(WARP_SIZE, 1, 1);
rms_norm_f32<WARP_SIZE><<<nrows, block_dims, 0, stream>>>(x, dst, ncols, eps);
rms_norm_f32<WARP_SIZE><<<blocks_num, block_dims, 0, stream>>>(x, dst, ncols, stride_row, stride_channel, stride_sample, eps);
} else {
const dim3 block_dims(1024, 1, 1);
rms_norm_f32<1024><<<nrows, block_dims, 0, stream>>>(x, dst, ncols, eps);
rms_norm_f32<1024><<<blocks_num, block_dims, 0, stream>>>(x, dst, ncols, stride_row, stride_channel, stride_sample, eps);
}
}
@ -233,19 +254,22 @@ void ggml_cuda_op_norm(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
float * dst_d = (float *) dst->data;
cudaStream_t stream = ctx.stream();
GGML_ASSERT(ggml_is_contiguous(src0));
GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32);
const int64_t ne00 = src0->ne[0];
const int64_t nrows = ggml_nrows(src0);
GGML_TENSOR_UNARY_OP_LOCALS;
float eps;
memcpy(&eps, dst->op_params, sizeof(float));
GGML_ASSERT(eps >= 0.0f);
norm_f32_cuda(src0_d, dst_d, ne00, nrows, eps, stream);
const size_t ts0 = ggml_type_size(src0->type);
GGML_ASSERT(nb00 == ts0);
const int64_t s01 = nb01 / ts0;
const int64_t s02 = nb02 / ts0;
const int64_t s03 = nb03 / ts0;
norm_f32_cuda(src0_d, dst_d, ne00, ne01, ne02, ne03, s01, s02, s03, eps, stream);
}
void ggml_cuda_op_group_norm(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
@ -254,8 +278,6 @@ void ggml_cuda_op_group_norm(ggml_backend_cuda_context & ctx, ggml_tensor * dst)
float * dst_d = (float *)dst->data;
cudaStream_t stream = ctx.stream();
GGML_ASSERT(ggml_is_contiguous(src0));
GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32);
@ -275,19 +297,22 @@ void ggml_cuda_op_rms_norm(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
float * dst_d = (float *) dst->data;
cudaStream_t stream = ctx.stream();
GGML_ASSERT(ggml_is_contiguous(src0));
GGML_ASSERT(src0->type == GGML_TYPE_F32);
GGML_ASSERT( dst->type == GGML_TYPE_F32);
const int64_t ne00 = src0->ne[0];
const int64_t nrows = ggml_nrows(src0);
GGML_TENSOR_UNARY_OP_LOCALS;
float eps;
memcpy(&eps, dst->op_params, sizeof(float));
GGML_ASSERT(eps >= 0.0f);
rms_norm_f32_cuda(src0_d, dst_d, ne00, nrows, eps, stream);
const size_t ts0 = ggml_type_size(src0->type);
GGML_ASSERT(nb00 == ts0);
const int64_t s01 = nb01 / ts0;
const int64_t s02 = nb02 / ts0;
const int64_t s03 = nb03 / ts0;
rms_norm_f32_cuda(src0_d, dst_d, ne00, ne01, ne02, ne03, s01, s02, s03, eps, stream);
}
void ggml_cuda_op_rms_norm_back(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {

4
ggml/src/ggml-cuda/softmax.cu
View file

@ -18,7 +18,7 @@ __device__ float __forceinline__ t2f32<half>(half val) {
#ifdef __clang__
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wpass-failed"
#endif
#endif // __clang__
template <bool use_shared, int ncols_template, int block_size_template, typename T>
static __global__ void soft_max_f32(
const float * x, const T * mask, float * dst, const int ncols_par, const int nrows_y,
@ -126,7 +126,7 @@ static __global__ void soft_max_f32(
}
#ifdef __clang__
#pragma clang diagnostic pop
#endif
#endif // __clang__
static __global__ void soft_max_back_f32(
const float * grad, const float * dstf, float * dst, const int ncols, const float scale) {

10
ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-cpb16.cu Normal file
View file

@ -0,0 +1,10 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-mma-f16.cuh"
DECL_FATTN_MMA_F16_CASE(64, 16);
DECL_FATTN_MMA_F16_CASE(80, 16);
DECL_FATTN_MMA_F16_CASE(96, 16);
DECL_FATTN_MMA_F16_CASE(112, 16);
DECL_FATTN_MMA_F16_CASE(128, 16);
DECL_FATTN_MMA_F16_CASE(256, 16);

10
ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-cpb32.cu Normal file
View file

@ -0,0 +1,10 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-mma-f16.cuh"
DECL_FATTN_MMA_F16_CASE(64, 32);
DECL_FATTN_MMA_F16_CASE(80, 32);
DECL_FATTN_MMA_F16_CASE(96, 32);
DECL_FATTN_MMA_F16_CASE(112, 32);
DECL_FATTN_MMA_F16_CASE(128, 32);
DECL_FATTN_MMA_F16_CASE(256, 32);

10
ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-cpb64.cu Normal file
View file

@ -0,0 +1,10 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-mma-f16.cuh"
DECL_FATTN_MMA_F16_CASE(64, 64);
DECL_FATTN_MMA_F16_CASE(80, 64);
DECL_FATTN_MMA_F16_CASE(96, 64);
DECL_FATTN_MMA_F16_CASE(112, 64);
DECL_FATTN_MMA_F16_CASE(128, 64);
DECL_FATTN_MMA_F16_CASE(256, 64);

10
ggml/src/ggml-cuda/template-instances/fattn-mma-f16-instance-cpb8.cu Normal file
View file

@ -0,0 +1,10 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-mma-f16.cuh"
DECL_FATTN_MMA_F16_CASE(64, 8);
DECL_FATTN_MMA_F16_CASE(80, 8);
DECL_FATTN_MMA_F16_CASE(96, 8);
DECL_FATTN_MMA_F16_CASE(112, 8);
DECL_FATTN_MMA_F16_CASE(128, 8);
DECL_FATTN_MMA_F16_CASE(256, 8);

10
ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb16.cu
View file

@ -1,10 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
DECL_FATTN_WMMA_F16_CASE(64, 16, float);
DECL_FATTN_WMMA_F16_CASE(80, 16, float);
DECL_FATTN_WMMA_F16_CASE(96, 16, float);
DECL_FATTN_WMMA_F16_CASE(112, 16, float);
DECL_FATTN_WMMA_F16_CASE(128, 16, float);
DECL_FATTN_WMMA_F16_CASE(256, 16, float);

9
ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqfloat-cpb32.cu
View file

@ -1,9 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
DECL_FATTN_WMMA_F16_CASE(64, 32, float);
DECL_FATTN_WMMA_F16_CASE(80, 32, float);
DECL_FATTN_WMMA_F16_CASE(96, 32, float);
DECL_FATTN_WMMA_F16_CASE(112, 32, float);
DECL_FATTN_WMMA_F16_CASE(128, 32, float);

10
ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb16.cu
View file

@ -1,10 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
DECL_FATTN_WMMA_F16_CASE(64, 16, half);
DECL_FATTN_WMMA_F16_CASE(80, 16, half);
DECL_FATTN_WMMA_F16_CASE(96, 16, half);
DECL_FATTN_WMMA_F16_CASE(112, 16, half);
DECL_FATTN_WMMA_F16_CASE(128, 16, half);
DECL_FATTN_WMMA_F16_CASE(256, 16, half);

10
ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb32.cu
View file

@ -1,10 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
DECL_FATTN_WMMA_F16_CASE(64, 32, half);
DECL_FATTN_WMMA_F16_CASE(80, 32, half);
DECL_FATTN_WMMA_F16_CASE(96, 32, half);
DECL_FATTN_WMMA_F16_CASE(112, 32, half);
DECL_FATTN_WMMA_F16_CASE(128, 32, half);
DECL_FATTN_WMMA_F16_CASE(256, 32, half);

8
ggml/src/ggml-cuda/template-instances/fattn-wmma-f16-instance-kqhalf-cpb8.cu
View file

@ -1,8 +0,0 @@
// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
DECL_FATTN_WMMA_F16_CASE(64, 8, half);
DECL_FATTN_WMMA_F16_CASE(96, 8, half);
DECL_FATTN_WMMA_F16_CASE(128, 8, half);
DECL_FATTN_WMMA_F16_CASE(256, 8, half);

22
ggml/src/ggml-cuda/template-instances/generate_cu_files.py
View file

@ -12,13 +12,13 @@ SOURCE_FATTN_VEC = """// This file has been autogenerated by generate_cu_files.p
DECL_FATTN_VEC_F{vkq_size}_CASE({head_size}, {type_k}, {type_v});
"""
SOURCE_FATTN_WMMA_START = """// This file has been autogenerated by generate_cu_files.py, do not edit manually.
SOURCE_FATTN_MMA_START = """// This file has been autogenerated by generate_cu_files.py, do not edit manually.
#include "../fattn-wmma-f16.cuh"
#include "../fattn-mma-f16.cuh"
"""
SOURCE_FATTN_WMMA_CASE = "DECL_FATTN_WMMA_F16_CASE({head_size}, {cols_per_block}, {kq_acc_t});\n"
SOURCE_FATTN_MMA_CASE = "DECL_FATTN_MMA_F16_CASE({head_size}, {cols_per_block});\n"
TYPES_MMQ = [
"GGML_TYPE_Q4_0", "GGML_TYPE_Q4_1", "GGML_TYPE_Q5_0", "GGML_TYPE_Q5_1", "GGML_TYPE_Q8_0",
@ -57,20 +57,12 @@ for vkq_size in [16, 32]:
with open(f"fattn-vec-f{vkq_size}-instance-hs{head_size}-{get_short_name(type_k)}-{get_short_name(type_v)}.cu", "w") as f:
f.write(SOURCE_FATTN_VEC.format(vkq_size=vkq_size, head_size=head_size, type_k=type_k, type_v=type_v))
for kq_acc_t in ["half", "float"]:
for cols_per_block in [8, 16, 32]:
if kq_acc_t == "float" and cols_per_block == 8:
continue
with open(f"fattn-wmma-f16-instance-kq{kq_acc_t}-cpb{cols_per_block}.cu", "w") as f:
f.write(SOURCE_FATTN_WMMA_START)
for cols_per_block in [8, 16, 32, 64]:
with open(f"fattn-mma-f16-instance-cpb{cols_per_block}.cu", "w") as f:
f.write(SOURCE_FATTN_MMA_START)
for head_size in [64, 80, 96, 112, 128, 256]:
if cols_per_block == 8 and head_size % 32 != 0: # wmma fragment is 8x32
continue
if kq_acc_t == "float" and cols_per_block == 32 and head_size == 256: # register spilling, bad performance
continue
f.write(SOURCE_FATTN_WMMA_CASE.format(kq_acc_t=kq_acc_t, cols_per_block=cols_per_block, head_size=head_size))
f.write(SOURCE_FATTN_MMA_CASE.format(cols_per_block=cols_per_block, head_size=head_size))
for type in TYPES_MMQ:
with open(f"mmq-instance-{get_short_name(type)}.cu", "w") as f:

3
ggml/src/ggml-cuda/vendors/hip.h vendored
View file

@ -1,5 +1,6 @@
#pragma once
#define HIP_ENABLE_WARP_SYNC_BUILTINS 1
#include <hip/hip_runtime.h>
#include <hipblas/hipblas.h>
#include <hip/hip_fp16.h>
@ -8,6 +9,7 @@
// for rocblas_initialize()
#include "rocblas/rocblas.h"
#endif // __HIP_PLATFORM_AMD__
#define CUBLAS_COMPUTE_16F HIPBLAS_R_16F
#define CUBLAS_COMPUTE_32F HIPBLAS_R_32F
#define CUBLAS_COMPUTE_32F_FAST_16F HIPBLAS_R_32F
@ -25,6 +27,7 @@
#define CU_MEM_LOCATION_TYPE_DEVICE hipMemLocationTypeDevice
#define CU_MEM_ACCESS_FLAGS_PROT_READWRITE hipMemAccessFlagsProtReadWrite
#define CU_CHECK(fn) {hipError_t err = fn; if(err != hipSuccess) { GGML_ABORT("HipVMM Failure: %s\n", hipGetErrorString(err)); }}
#define __shfl_sync(mask, var, laneMask, width) __shfl(var, laneMask, width)
#define __shfl_xor_sync(mask, var, laneMask, width) __shfl_xor(var, laneMask, width)
#define cublasComputeType_t hipblasDatatype_t //deprecated, new hipblasComputeType_t not in 5.6
#define cublasCreate hipblasCreate

20
ggml/src/ggml-metal/ggml-metal.m
View file

@ -19,8 +19,15 @@
// max number of MTLCommandBuffer used to submit a graph for processing
#define GGML_METAL_MAX_COMMAND_BUFFERS 8
#ifndef TARGET_OS_VISION
#define TARGET_OS_VISION 0
#endif
// create residency sets only on macOS >= 15.0
#if TARGET_OS_OSX && __MAC_OS_X_VERSION_MAX_ALLOWED >= 150000
#if TARGET_OS_OSX && __MAC_OS_X_VERSION_MAX_ALLOWED >= 150000 || \
TARGET_OS_IOS && __IPHONE_OS_VERSION_MAX_ALLOWED >= 180000 || \
TARGET_OS_TV && __TV_OS_VERSION_MAX_ALLOWED >= 180000 || \
TARGET_OS_VISION && __VISION_OS_VERSION_MAX_ALLOWED >= 200000
#define GGML_METAL_HAS_RESIDENCY_SETS 1
#endif
@ -1071,7 +1078,7 @@ static bool ggml_backend_metal_buffer_rset_init(
}
#if defined(GGML_METAL_HAS_RESIDENCY_SETS)
if (@available(macOS 15.0, *)) {
if (@available(macOS 15.0, iOS 18.0, tvOS 18.0, visionOS 2.0, *)) {
MTLResidencySetDescriptor * desc = [[MTLResidencySetDescriptor alloc] init];
desc.label = @"ggml_backend_metal";
desc.initialCapacity = ctx->n_buffers;
@ -1106,7 +1113,7 @@ static bool ggml_backend_metal_buffer_rset_init(
// rset free
static void ggml_backend_metal_buffer_rset_free(struct ggml_backend_metal_buffer_context * ctx) {
#if defined(GGML_METAL_HAS_RESIDENCY_SETS)
if (@available(macOS 15.0, *)) {
if (@available(macOS 15.0, iOS 18.0, tvOS 18.0, visionOS 2.0, *)) {
if (ctx->rset) {
[ctx->rset endResidency];
[ctx->rset removeAllAllocations];
@ -1201,12 +1208,13 @@ static bool ggml_metal_supports_op(const struct ggml_backend_metal_device_contex
case GGML_OP_SUM_ROWS:
case GGML_OP_SOFT_MAX:
case GGML_OP_GROUP_NORM:
return has_simdgroup_reduction;
return has_simdgroup_reduction && ggml_is_contiguous(op->src[0]);
case GGML_OP_RMS_NORM:
return has_simdgroup_reduction && (op->ne[0] % 4 == 0);
return has_simdgroup_reduction && (op->ne[0] % 4 == 0 && ggml_is_contiguous_1(op->src[0]));
case GGML_OP_ARGMAX:
case GGML_OP_NORM:
return true;
case GGML_OP_NORM:
return has_simdgroup_reduction && (op->ne[0] % 4 == 0 && ggml_is_contiguous_1(op->src[0]));
case GGML_OP_ROPE:
{
const int mode = ((const int32_t *) op->op_params)[2];

6
ggml/src/ggml-sycl/ggml-sycl.cpp
View file

@ -4537,14 +4537,17 @@ static bool ggml_backend_sycl_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_VIEW:
case GGML_OP_PERMUTE:
case GGML_OP_TRANSPOSE:
case GGML_OP_NORM:
case GGML_OP_ADD:
case GGML_OP_ADD1:
case GGML_OP_LOG:
case GGML_OP_SUB:
case GGML_OP_MUL:
case GGML_OP_DIV:
return true;
case GGML_OP_NORM:
case GGML_OP_RMS_NORM:
case GGML_OP_GROUP_NORM:
return ggml_is_contiguous(op->src[0]);
case GGML_OP_SCALE:
case GGML_OP_SQR:
case GGML_OP_SQRT:
@ -4576,7 +4579,6 @@ static bool ggml_backend_sycl_device_supports_op(ggml_backend_dev_t dev, const g
case GGML_OP_SUM_ROWS:
case GGML_OP_ARGSORT:
case GGML_OP_ACC:
case GGML_OP_GROUP_NORM:
case GGML_OP_UPSCALE:
case GGML_OP_PAD:
case GGML_OP_LEAKY_RELU:

62935
ggml/src/ggml-vulkan-shaders.cpp
View file

File diff suppressed because it is too large Load diff

Some files were not shown because too many files have changed in this diff Show more