vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Add doc and better string handling

Browse source
This commit is contained in:
Bjarke Viksøe 2024-07-06 21:08:53 +02:00
parent a21b89fd0e
commit bc3ed77ddf
2 changed files with 5 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

4
examples/server/README.md
View file

@ -366,7 +366,8 @@ Notice that each `probs` is an array of length `n_probs`.
"assistant_name": "", "assistant_name": "",
"user_name": "", "user_name": "",
"default_generation_settings": { ... }, "default_generation_settings": { ... },
"total_slots": 1 "total_slots": 1,
"model_template": ""
} }
``` ```
@ -374,6 +375,7 @@ Notice that each `probs` is an array of length `n_probs`.
- `user_name` - the required anti-prompt to generate the prompt in case you have specified a system prompt for all slots. - `user_name` - the required anti-prompt to generate the prompt in case you have specified a system prompt for all slots.
- `default_generation_settings` - the default generation settings for the `/completion` endpoint, which has the same fields as the `generation_settings` response object from the `/completion` endpoint. - `default_generation_settings` - the default generation settings for the `/completion` endpoint, which has the same fields as the `generation_settings` response object from the `/completion` endpoint.
- `total_slots` - the total number of slots for process requests (defined by `--parallel` option) - `total_slots` - the total number of slots for process requests (defined by `--parallel` option)
- `model_template` - the model's original Jinja2 prompt template
- **POST** `/v1/chat/completions`: OpenAI-compatible Chat Completions API. Given a ChatML-formatted json description in `messages`, it returns the predicted completion. Both synchronous and streaming mode are supported, so scripted and interactive applications work fine. While no strong claims of compatibility with OpenAI API spec is being made, in our experience it suffices to support many apps. Only models with a [supported chat template](https://github.com/ggerganov/llama.cpp/wiki/Templates-supported-by-llama_chat_apply_template) can be used optimally with this endpoint. By default, the ChatML template will be used. - **POST** `/v1/chat/completions`: OpenAI-compatible Chat Completions API. Given a ChatML-formatted json description in `messages`, it returns the predicted completion. Both synchronous and streaming mode are supported, so scripted and interactive applications work fine. While no strong claims of compatibility with OpenAI API spec is being made, in our experience it suffices to support many apps. Only models with a [supported chat template](https://github.com/ggerganov/llama.cpp/wiki/Templates-supported-by-llama_chat_apply_template) can be used optimally with this endpoint. By default, the ChatML template will be used.

6
examples/server/server.cpp
View file

@ -2970,10 +2970,8 @@ int main(int argc, char ** argv) {
std::string template_key = "tokenizer.chat_template", curr_tmpl; std::string template_key = "tokenizer.chat_template", curr_tmpl;
int32_t tlen = llama_model_meta_val_str(ctx_server.model, template_key.c_str(), nullptr, 0); int32_t tlen = llama_model_meta_val_str(ctx_server.model, template_key.c_str(), nullptr, 0);
if (tlen > 0) { if (tlen > 0) {
std::vector<char> model_template(tlen + 1, 0); curr_tmpl.resize(tlen + 1);
if (llama_model_meta_val_str(ctx_server.model, template_key.c_str(), model_template.data(), model_template.size()) > 0) { llama_model_meta_val_str(ctx_server.model, template_key.c_str(), &curr_tmpl[0], curr_tmpl.size());
curr_tmpl = std::string(model_template.data(), model_template.size());
}
} }
res.set_header("Access-Control-Allow-Origin", req.get_header_value("Origin")); res.set_header("Access-Control-Allow-Origin", req.get_header_value("Origin"));
json data = { json data = {