tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-28 20:25:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

server: add --reasoning-budget 0 to disable thinking (incl. qwen3 w/ enable_thinking:false) (#13771)

Browse Source 
---------

Co-authored-by: ochafik <ochafik@google.com>
Co-authored-by: Xuan-Son Nguyen <thichthat@gmail.com>
This commit is contained in:
Olivier Chafik
2025-05-26 00:30:51 +01:00
committed by GitHub
parent 2f099b510f
commit e121edc432
12 changed files with 277 additions and 107 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
tools/server/server.cpp
View File

@ -178,7 +178,7 @@ struct slot_params {
{"grammar_triggers", grammar_triggers},
{"preserved_tokens", sampling.preserved_tokens},
{"chat_format", common_chat_format_name(oaicompat_chat_syntax.format)},
{"reasoning_format", (oaicompat_chat_syntax.reasoning_format == COMMON_REASONING_FORMAT_DEEPSEEK ? "deepseek" : "none")},
{"reasoning_format", common_reasoning_format_name(oaicompat_chat_syntax.reasoning_format)},
{"reasoning_in_content", oaicompat_chat_syntax.reasoning_in_content},
{"thinking_forced_open", oaicompat_chat_syntax.thinking_forced_open},
{"samplers", samplers},
@ -357,7 +357,7 @@ struct server_task {
auto it = data.find("chat_format");
if (it != data.end()) {
params.oaicompat_chat_syntax.format = static_cast<common_chat_format>(it->get<int>());
SRV_INF("Chat format: %s\n", common_chat_format_name(params.oaicompat_chat_syntax.format).c_str());
SRV_INF("Chat format: %s\n", common_chat_format_name(params.oaicompat_chat_syntax.format));
} else {
params.oaicompat_chat_syntax.format = defaults.oaicompat_chat_syntax.format;
}
@ -2089,6 +2089,7 @@ struct server_context {
/* common_chat_templates */ chat_templates.get(),
/* allow_image */ mctx ? mtmd_support_vision(mctx) : false,
/* allow_audio */ mctx ? mtmd_support_audio (mctx) : false,
/* enable_thinking */ params_base.reasoning_budget != 0,
};
}