thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

server : refactor oai_parser_opt, move it to server_chat_params (#18937)

Browse source 
* server_chat_params

* move chat format into CLI

* use meta whenever possible

* clean up, no more chatml fallback
This commit is contained in:
Xuan-Son Nguyen 2026-01-19 23:28:01 +01:00 • committed by GitHub
parent 1706a6d7c6
commit 6df686bee6
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
8 changed files with 112 additions and 103 deletions
Download patch file Download diff file Expand all files Collapse all files

5
tools/server/server-context.h
View file

@ -20,9 +20,8 @@ struct server_context_meta {
int slot_n_ctx;
enum llama_pooling_type pooling_type;
// chat template
std::string chat_template;
std::string chat_template_tool_use;
// chat params
server_chat_params & chat_params;
// tokens
std::string bos_token_str;