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

server : clean up built-in template detection (#11026)

Browse source 
* server : clean up built-in template detection

* fix compilation

* add chat template test

* fix condition
This commit is contained in:
Xuan Son Nguyen 2024-12-31 15:22:01 +01:00 • committed by GitHub
parent 5896c65232
commit 45095a61bf
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
6 changed files with 44 additions and 27 deletions
Download patch file Download diff file Expand all files Collapse all files

12
common/common.cpp
View file

@ -1614,6 +1614,18 @@ std::string common_detokenize(llama_context * ctx, const std::vector<llama_token
// Chat template utils
//
std::string common_get_builtin_chat_template(const struct llama_model * model) {
static const char * template_key = "tokenizer.chat_template";
// call with NULL buffer to get the total size of the string
int32_t res = llama_model_meta_val_str(model, template_key, NULL, 0);
if (res > 0) {
std::vector<char> model_template(res + 1, 0);
llama_model_meta_val_str(model, template_key, model_template.data(), model_template.size());
return std::string(model_template.data(), model_template.size() - 1);
}
return "";
}
bool common_chat_verify_template(const std::string & tmpl) {
llama_chat_message chat[] = {{"user", "test"}};
int res = llama_chat_apply_template(nullptr, tmpl.c_str(), chat, 1, true, nullptr, 0);

3
common/common.h
View file

@ -571,6 +571,9 @@ struct common_chat_msg {
std::string content;
};
// Get the built-in chat template for the model. Return empty string if not present.
std::string common_get_builtin_chat_template(const struct llama_model * model);
// Check if the template supplied via "--chat-template" is supported or not. Returns true if it's valid
bool common_chat_verify_template(const std::string & tmpl);