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

server : support preserving reasoning_content in assistant message (#18994)

Browse source 
* support reasoning_content input

* report template caps to webui

* add docs

* rm commented code
This commit is contained in:
Xuan-Son Nguyen 2026-01-22 21:30:06 +01:00 • committed by GitHub
parent a5eaa1d6a3
commit 51fa458a92
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 165 additions and 131 deletions
Download patch file Download diff file Expand all files Collapse all files

2
tools/server/server-context.cpp
View file

@ -2903,6 +2903,7 @@ server_context_meta server_context::get_meta() const {
/* pooling_type */ llama_pooling_type(impl->ctx),
/* chat_params */ impl->chat_params,
/* chat_template_caps */ common_chat_templates_get_caps(impl->chat_params.tmpls.get()),
/* bos_token_str */ bos_token_str,
/* eos_token_str */ eos_token_str,
@ -3410,6 +3411,7 @@ void server_routes::init_routes() {
{ "webui", params.webui },
{ "webui_settings", meta->json_webui_settings },
{ "chat_template", tmpl_default },
{ "chat_template_caps", meta->chat_template_caps },
{ "bos_token", meta->bos_token_str },
{ "eos_token", meta->eos_token_str },
{ "build_info", meta->build_info },