Update environment configuration to switch OpenAI model from gpt-5.4 to gpt-4o-mini. Introduce _effective_llm_route_from_request function in api_server.py to determine the effective LLM routing based on service keys and model compatibility. Update chat event handling to include LLM routing information in response data. Enhance logging for better traceability of API interactions.

2026-04-16 15:49:03 +08:00
parent be15b512a4
commit 70c8e90f2e
4 changed files with 104 additions and 2 deletions
@@ -13,8 +13,8 @@ MODEL_PRIMARY=deepseek-chat

 OPENAI_API_KEY=sk-proj-FEb2ChHZK5Llm0tBkmNIT3BlbkFJbu0b0zQpTW8762yD6HDv
 OPENAI_BASE_URL=http://113.192.49.54:9080/v1
-OPENAI_MODEL=gpt-5.4
-# OPENAI_CHAT_MODEL=gpt-4o-mini
+OPENAI_MODEL=gpt-4o-mini
+# OPENAI_CHAT_MODEL=

 # 模型配置
 # 主生成模型