LLM inference in C/C++
C++ ggml
server : fix reasoning budget WebUI precedence over model.ini (#24517)
Georgi Gerganovebc10770ac5a
diff --git a/tools/server/server-common.cpp b/tools/server/server-common.cppindex 9f3caac8f7..aebaf1f91b 100644--- a/tools/server/server-common.cpp+++ b/tools/server/server-common.cpp@@ -1126,9 +1126,9 @@ json oaicompat_chat_params_parse(// Reasoning budget: pass parameters through to sampling layer{- int reasoning_budget = opt.reasoning_budget;- if (reasoning_budget == -1 && body.contains("thinking_budget_tokens")) {- reasoning_budget = json_value(body, "thinking_budget_tokens", -1);+ int reasoning_budget = json_value(body, "thinking_budget_tokens", -1);+ if (reasoning_budget == -1) {+ reasoning_budget = opt.reasoning_budget;}if (!chat_params.thinking_end_tag.empty()) {