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

server : allow specifying reasoning_format in HTTP request (#15238)

Browse source
This commit is contained in:
Xuan-Son Nguyen 2025-08-11 14:48:41 +02:00 • committed by GitHub
parent 27093afe78
commit 53d0a12658
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
7 changed files with 28 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

6
common/arg.cpp
View file

@ -2949,11 +2949,7 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
"- deepseek: puts thoughts in `message.reasoning_content` (except in streaming mode, which behaves as `none`)\n"
"(default: auto)",
[](common_params & params, const std::string & value) {
/**/ if (value == "deepseek") { params.reasoning_format = COMMON_REASONING_FORMAT_DEEPSEEK; }
else if (value == "deepseek-legacy") { params.reasoning_format = COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY; }
else if (value == "none") { params.reasoning_format = COMMON_REASONING_FORMAT_NONE; }
else if (value == "auto") { params.reasoning_format = COMMON_REASONING_FORMAT_AUTO; }
else { throw std::invalid_argument("invalid value"); }
params.reasoning_format = common_reasoning_format_from_name(value);
}
).set_examples({LLAMA_EXAMPLE_SERVER, LLAMA_EXAMPLE_MAIN}).set_env("LLAMA_ARG_THINK"));
add_opt(common_arg(

13
common/chat.cpp
View file

@ -625,6 +625,19 @@ const char * common_reasoning_format_name(common_reasoning_format format) {
}
}
common_reasoning_format common_reasoning_format_from_name(const std::string & format) {
if (format == "none") {
return COMMON_REASONING_FORMAT_NONE;
} else if (format == "auto") {
return COMMON_REASONING_FORMAT_AUTO;
} else if (format == "deepseek") {
return COMMON_REASONING_FORMAT_DEEPSEEK;
} else if (format == "deepseek-legacy") {
return COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY;
}
throw std::runtime_error("Unknown reasoning format: " + format);
}
static std::string wrap_code_as_arguments(common_chat_msg_parser & builder, const std::string & code) {
std::string arguments;
if (builder.is_partial()) {

1
common/chat.h
View file

@ -191,6 +191,7 @@ std::string common_chat_format_example(
const char* common_chat_format_name(common_chat_format format);
const char* common_reasoning_format_name(common_reasoning_format format);
common_reasoning_format common_reasoning_format_from_name(const std::string & format);
common_chat_msg common_chat_parse(const std::string & input, bool is_partial, const common_chat_syntax & syntax);
common_chat_tool_choice common_chat_tool_choice_parse_oaicompat(const std::string & tool_choice);

6
tools/server/README.md
View file

@ -1132,6 +1132,12 @@ The `response_format` parameter supports both plain JSON output (e.g. `{"type":
`chat_template_kwargs`: Allows sending additional parameters to the json templating system. For example: `{"enable_thinking": false}`
`reasoning_format`: The reasoning format to be parsed. If set to `none`, it will output the raw generated text.
`thinking_forced_open`: Force a reasoning model to always output the reasoning. Only works on certain models.
`parse_tool_calls`: Whether to parse the generated tool call.
*Examples:*
You can use either Python `openai` library with appropriate checkpoints:

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

Binary file not shown.

8
tools/server/server.cpp
View file

@ -383,8 +383,12 @@ struct server_task {
} else {
params.oaicompat_chat_syntax.format = defaults.oaicompat_chat_syntax.format;
}
params.oaicompat_chat_syntax.reasoning_format = params_base.reasoning_format;
params.oaicompat_chat_syntax.reasoning_in_content = params.stream && (params_base.reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY);
common_reasoning_format reasoning_format = params_base.reasoning_format;
if (data.contains("reasoning_format")) {
reasoning_format = common_reasoning_format_from_name(data.at("reasoning_format").get<std::string>());
}
params.oaicompat_chat_syntax.reasoning_format = reasoning_format;
params.oaicompat_chat_syntax.reasoning_in_content = params.stream && (reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK_LEGACY);
params.oaicompat_chat_syntax.thinking_forced_open = json_value(data, "thinking_forced_open", false);
params.oaicompat_chat_syntax.parse_tool_calls = json_value(data, "parse_tool_calls", false);
}

1
tools/server/webui/src/utils/app.context.tsx
View file

@ -209,6 +209,7 @@ export const AppContextProvider = ({
messages,
stream: true,
cache_prompt: true,
reasoning_format: 'none',
samplers: config.samplers,
temperature: config.temperature,
dynatemp_range: config.dynatemp_range,