1.修改部分火山模型前缀为volcengine，2.火山模型添加额外请求头，避免复用导致长连接僵化

config.yaml

@@ -11,13 +11,15 @@ model_list:
   - model_name: deepseek-v3.2
     litellm_params:
       <<: *token_pricing
-      model:  openai/deepseek-v3-2-251201
+      model:  volcengine/deepseek-v3-2-251201
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
       # 说明中建议设置30分钟(1800秒)以上，避免深度思考过程中断连接
       timeout: 1800       
       # 关键配置2：启用思考模式
       # extra_body 中的参数会被直接合并到请求的 JSON body 中
+      extra_headers:
+        Connection: "close"
       extra_body:
         thinking:
           type: "enabled"
@@ -25,9 +27,11 @@ model_list:
   - model_name: deepseek-v3.2-nothink
     litellm_params:
       <<: *token_pricing
-      model:  openai/deepseek-v3-2-251201
+      model:  volcengine/deepseek-v3-2-251201
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY
+      extra_headers:
+        Connection: "close" 
       extra_body:
         thinking:
           type: "disabled" 
@@ -36,32 +40,40 @@ model_list:
   - model_name: doubao-seed-1-8
     litellm_params:
       <<: *token_pricing
-      model: openai/doubao-seed-1-8-251228
+      model: volcengine/doubao-seed-1-8-251228
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY
+      extra_headers:
+        Connection: "close" 
 
   - model_name: doubao-seed-1-8-nothink
     litellm_params:
       <<: *token_pricing
-      model: openai/doubao-seed-1-8-251228
+      model: volcengine/doubao-seed-1-8-251228
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY
+      extra_headers:
+        Connection: "close" 
       extra_body:
         reasoning_effort: "minimal"        
 
   - model_name: glm-4.7
     litellm_params:
       <<: *token_pricing
-      model: openai/glm-4-7-251222
+      model: volcengine/glm-4-7-251222
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
+      extra_headers:
+        Connection: "close"
 
   - model_name: glm-4.7-nothink
     litellm_params:
       <<: *token_pricing
-      model: openai/glm-4-7-251222
+      model: volcengine/glm-4-7-251222
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
+      extra_headers:
+        Connection: "close"
       extra_body:
         thinking:
           type: "disabled" 
@@ -78,9 +90,11 @@ model_list:
   - model_name: test-ggllmm
     litellm_params:
       <<: *token_pricing
-      model: openai/glm-4-7-251222
+      model: volcengine/glm-4-7-251222
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
+      extra_headers:
+        Connection: "close"
       extra_body:
         thinking:
           type: "disabled" 
@@ -88,9 +102,11 @@ model_list:
   - model_name: test-dbao
     litellm_params:
       <<: *token_pricing
-      model: openai/doubao-seed-1-8-251228
+      model: volcengine/doubao-seed-1-8-251228
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
+      extra_headers:
+        Connection: "close"      
       extra_body:
         reasoning_effort: "minimal" 
 
@@ -102,6 +118,8 @@ model_list:
       api_key: os.environ/HUOSHAN_API_KEY 
       timeout: 120 
       stream: false 
+      extra_headers:
+        Connection: "close"               
       extra_body:
         reasoning_effort: "minimal"    
 
@@ -114,6 +132,8 @@ model_list:
       api_key: os.environ/HUOSHAN_API_KEY 
       timeout: 120 
       stream: false 
+      extra_headers:
+        Connection: "close"                    
       extra_body:
         thinking:
           type: "disabled" 
@@ -129,7 +149,8 @@ model_list:
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY 
       timeout: 1800           
-
+      extra_headers:
+        Connection: "close"