修改config，添加模型；指定litellm主版本规避风险

config.yaml

@@ -64,6 +64,39 @@ model_list:
       extra_body:
         reasoning_effort: "minimal"        
 
+
+  - model_name: doubao-seed-2-0-pro
+    litellm_params:
+      <<: *token_pricing
+      model: volcengine/doubao-seed-2-0-pro-260215
+      api_base: os.environ/HUOSHAN_BASE_URL
+      api_key: os.environ/HUOSHAN_API_KEY
+      timeout: 1800  
+      extra_headers:
+        Connection: "close"
+
+  # doubao-seed-2-0-pro 强制中等思考，用作回退
+  - model_name: doubao-seed-2-0-pro-medium
+    litellm_params:
+      <<: *token_pricing
+      model: volcengine/doubao-seed-2-0-pro-260215
+      api_base: os.environ/HUOSHAN_BASE_URL
+      api_key: os.environ/HUOSHAN_API_KEY
+      timeout: 1800  
+      extra_headers:
+        Connection: "close"
+      extra_body:
+        reasoning_effort: "medium"
+        
+  - model_name: doubao-seed-2-0-lite
+    litellm_params:
+      <<: *token_pricing
+      model: volcengine/doubao-seed-2-0-lite-260215
+      api_base: os.environ/HUOSHAN_BASE_URL
+      api_key: os.environ/HUOSHAN_API_KEY
+      extra_headers:
+        Connection: "close"
+
   - model_name: glm-4.7
     litellm_params:
       <<: *token_pricing
@@ -100,11 +133,10 @@ model_list:
       extra_body:
         reasoning_effort: "minimal"    
 
-          
   - model_name: fanyi
     litellm_params:
       <<: *token_pricing
-      model: volcengine/glm-4-7-251222
+      model: volcengine/doubao-seed-2-0-lite-260215
       api_base: os.environ/HUOSHAN_BASE_URL
       api_key: os.environ/HUOSHAN_API_KEY
       timeout: 120 
@@ -112,8 +144,9 @@ model_list:
       extra_headers:
         Connection: "close"               
       extra_body:
-        thinking:
+        reasoning_effort: "minimal"
-          type: "disabled" 
+
+          
 
 
 
@@ -124,7 +157,7 @@ litellm_settings:
 #  fallbacks: [
 #    {"gemini-2.5-flash": ["gemini-2.5-pro"]}
 #  ]
-#  default_fallbacks: ["gemini-2.5-flash"]
+  default_fallbacks: ["doubao-seed-2-0-pro-medium"]
   request_timeout: 120 # raise Timeout error if call takes longer than 10s. Sets litellm.request_timeout 
   cache: False          # 开启缓存
 #  cache_params:
@@ -132,7 +165,6 @@ litellm_settings:
 #    supported_call_types: ["completion", "embedding", "acompletion"] # 支持的类型  
   allowed_fails: 3 # cooldown model if it fails > 1 call in a minute.
   # 关键配置：注册自定义回调，实现单上游模型每日token限制的关键
-  #callbacks: ["daily_limit.proxy_handler_instance"]
   callbacks: ["volc_limit_sync.proxy_handler_instance"]
 
 router_settings: # router_settings are optional

daily_limit.py

@@ -48,9 +48,10 @@ REDIS_PASSWORD = os.getenv("REDIS_PASSWORD", None)
 
 MODEL_QUOTAS = {
     "deepseek-v3-2-251201:ba0383d3": 1600000,
-    "doubao-seed-1-8-251228:ba0383d3": 1800000,  # 测试限额
+    "doubao-seed-1-8-251228:ba0383d3": 1800000,
     "glm-4-7-251222:ba0383d3": 1800000,
-    "doubao-seed-1-8-251228:d57edef4": 1500000,  # 测试重名
+    "doubao-seed-2-0-pro-260215:ba0383d3": 1900000,
+    "doubao-seed-2-0-lite-260215:ba0383d3": 1900000,
 }
 
 

docker-compose.yml

@@ -1,7 +1,8 @@
 services:
   litellm:
     mem_limit: 2048m
-    image: ghcr.io/berriai/litellm:main-latest
+    #image: ghcr.io/berriai/litellm:main-latest # 注意：1.82.7，1.82.8被投毒
+    image: ghcr.io/berriai/litellm@sha256:7c311546c25e7bb6e8cafede9fcd3d0d622ac636b5c9418befaa32e85dfb0186
     container_name: litellm
     restart: unless-stopped
     volumes: