vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-12 01:54:37 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

Merge branch 'master' into concedo_experimental

Browse source 
# Conflicts:
#	.devops/nix/scope.nix
#	.github/workflows/nix-ci-aarch64.yml
#	.github/workflows/nix-ci.yml
#	README.md
#	scripts/sync-ggml.last
This commit is contained in:
Concedo 2024-02-24 18:30:51 +08:00
commit 359a14d3c2
10 changed files with 251 additions and 72 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/server/README.md
View file

@ -151,7 +151,7 @@ node index.js
`temperature`: Adjust the randomness of the generated text (default: 0.8).
`dynatemp_range`: Dynamic temperature range (default: 0.0, 0.0 = disabled).
`dynatemp_range`: Dynamic temperature range. The final temperature will be in the range of `[temperature - dynatemp_range; temperature + dynatemp_range]` (default: 0.0, 0.0 = disabled).
`dynatemp_exponent`: Dynamic temperature exponent (default: 1.0).
@ -209,7 +209,7 @@ node index.js
`slot_id`: Assign the completion task to an specific slot. If is -1 the task will be assigned to a Idle slot (default: -1)
`cache_prompt`: Save the prompt and generation for avoid reprocess entire prompt if a part of this isn't change (default: false)
`cache_prompt`: Re-use previously cached prompt from the last request if possible. This may prevent re-caching the prompt from scratch. (default: false)
`system_prompt`: Change the system prompt (initial prompt of all slots), this is useful for chat applications. [See more](#change-system-prompt-on-runtime)
@ -242,7 +242,7 @@ Notice that each `probs` is an array of length `n_probs`.
- `content`: Completion result as a string (excluding `stopping_word` if any). In case of streaming mode, will contain the next token as a string.
- `stop`: Boolean for use with `stream` to check whether the generation has stopped (Note: This is not related to stopping words array `stop` from input options)
- `generation_settings`: The provided options above excluding `prompt` but including `n_ctx`, `model`
- `generation_settings`: The provided options above excluding `prompt` but including `n_ctx`, `model`. These options may differ from the original ones in some way (e.g. bad values filtered out, strings converted to tokens, etc.).
- `model`: The path to the model loaded with `-m`
- `prompt`: The provided `prompt`
- `stopped_eos`: Indicating whether the completion has stopped because it encountered the EOS token

25
examples/server/server.cpp
View file

@ -401,6 +401,16 @@ struct llama_server_context
return true;
}
void validate_model_chat_template(server_params & sparams) {
llama_chat_message chat[] = {{"user", "test"}};
std::vector<char> buf(1);
int res = llama_chat_apply_template(model, nullptr, chat, 1, true, buf.data(), buf.size());
if (res < 0) {
LOG_ERROR("The chat template comes with this model is not yet supported, falling back to chatml. This may cause the model to output suboptimal responses", {});
sparams.chat_template = "<|im_start|>"; // llama_chat_apply_template only checks if <|im_start|> exist in the template
}
}
void initialize() {
// create slots
all_slots_are_idle = true;
@ -1939,6 +1949,10 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
printf(" -cb, --cont-batching enable continuous batching (a.k.a dynamic batching) (default: disabled)\n");
printf(" -spf FNAME, --system-prompt-file FNAME\n");
printf(" set a file to load a system prompt (initial prompt of all slots), this is useful for chat applications.\n");
printf(" -ctk TYPE, --cache-type-k TYPE\n");
printf(" KV cache data type for K (default: f16)\n");
printf(" -ctv TYPE, --cache-type-v TYPE\n");
printf(" KV cache data type for V (default: f16)\n");
printf(" --mmproj MMPROJ_FILE path to a multimodal projector file for LLaVA.\n");
printf(" --log-disable disables logging to a file.\n");
printf(" --slots-endpoint-disable disables slots monitoring endpoint.\n");
@ -2377,6 +2391,12 @@ static void server_params_parse(int argc, char **argv, server_params &sparams,
);
llama.process_system_prompt_data(json::parse(systm_content));
}
else if (arg == "-ctk" || arg == "--cache-type-k") {
params.cache_type_k = argv[++i];
}
else if (arg == "-ctv" || arg == "--cache-type-v") {
params.cache_type_v = argv[++i];
}
else if(arg == "--mmproj")
{
if (++i >= argc)
@ -2753,6 +2773,11 @@ int main(int argc, char **argv)
LOG_INFO("model loaded", {});
}
if (sparams.chat_template.empty()) { // custom chat template is not supplied
// check if the template comes with the model is supported by us
llama.validate_model_chat_template(sparams);
}
// Middleware for API key validation
auto validate_api_key = [&sparams](const httplib::Request &req, httplib::Response &res) -> bool {
// If API key is not set, skip validation