##yaml锚点方便配置。手动设定价格，方便ui中查看 x-definitions: &token_pricing input_cost_per_token: 1 output_cost_per_token: 1 model_list: #=======================火山平台，限时免费 #日常使用ds v3.2 ，强制think - model_name: deepseek-v3.2 litellm_params: <<: *token_pricing model: openai/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_body: thinking: type: "enabled" - model_name: deepseek-v3.2-nothink litellm_params: <<: *token_pricing model: openai/deepseek-v3-2-251201 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: thinking: type: "disabled" - model_name: doubao-seed-1-8 litellm_params: <<: *token_pricing model: openai/doubao-seed-1-8-251228 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY - model_name: doubao-seed-1-8-nothink litellm_params: <<: *token_pricing model: openai/doubao-seed-1-8-251228 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: reasoning_effort: "minimal" - model_name: glm-4.7 litellm_params: <<: *token_pricing model: openai/glm-4-7-251222 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY - model_name: glm-4.7-nothink litellm_params: <<: *token_pricing model: openai/glm-4-7-251222 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: thinking: type: "disabled" #=======================fanyi - model_name: test-newapi-dbao litellm_params: <<: *token_pricing model: openai/doubao-seed-1-8-251228 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: test-ggllmm litellm_params: <<: *token_pricing model: openai/glm-4-7-251222 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: thinking: type: "disabled" - model_name: test-dbao litellm_params: <<: *token_pricing model: openai/doubao-seed-1-8-251228 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: reasoning_effort: "minimal" - model_name: fanyi litellm_params: <<: *token_pricing model: openai/doubao-seed-1-8-251228 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: reasoning_effort: "minimal" - model_name: fanyi litellm_params: <<: *token_pricing model: openai/glm-4-7-251222 api_base: os.environ/HUOSHAN_BASE_URL api_key: os.environ/HUOSHAN_API_KEY extra_body: thinking: type: "disabled" #=======================Claude系列 #=======================Claude-Opus-v4.5系列 - model_name: Claude-Opus-v4.5 litellm_params: <<: *token_pricing model: openai/claude-opus-4-5-thinking api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Opus-v4.5 litellm_params: <<: *token_pricing model: openai/claude-opus-4-5 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Opus-v4.5 litellm_params: <<: *token_pricing model: openai/claude-4.5-opus api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Opus-v4.5 litellm_params: <<: *token_pricing model: openai/claude-opus-4-5-20251101 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Opus-v4.5 litellm_params: <<: *token_pricing model: openai/cursor2-claude-4.1-opus api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY #=======================Claude-Sonnet-v4.5系列 - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-sonnet-4-5-20250929 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-sonnet-4-5 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-4.5-sonnet-official-reverse api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-4.5-sonnet api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-4.5-sonnet-docs api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-4.5-sonnet-2cc api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/claude-sonnet-4-5-thinking api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: Claude-Sonnet-v4.5 litellm_params: <<: *token_pricing model: openai/cursor2-claude-4.5-sonnet api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY #=======================gpt-5系列 - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/cursor2-gpt-5 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/cursor2-gpt-5-codex api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.1 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.1-codex api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.1-codex-2cx api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.1-codex-max-2cx api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.2 api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY - model_name: gpt-5 litellm_params: <<: *token_pricing model: openai/gpt-5.2-2cx api_base: os.environ/NEWAPI_BASE_URL api_key: os.environ/NEWAPI_API_KEY litellm_settings: num_retries: 2 # retry call 3 times on each model_name (e.g. zephyr-beta) request_timeout: 120 # raise Timeout error if call takes longer than 10s. Sets litellm.request_timeout #fallbacks: [{"zephyr-beta": ["gpt-4o"]}] # fallback to gpt-4o if call fails num_retries #context_window_fallbacks: [{"zephyr-beta": ["gpt-3.5-turbo-16k"]}, {"gpt-4o": ["gpt-3.5-turbo-16k"]}] # fallback to gpt-3.5-turbo-16k if context window error cache: False # 开启缓存 # cache_params: # type: "redis" # 指定使用 Redis # 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 routing_strategy: simple-shuffle # Literal["simple-shuffle", "least-busy", "usage-based-routing","latency-based-routing"], default="simple-shuffle" #model_group_alias: {"gpt-4": "gpt-4o"} # all requests with `gpt-4` will be routed to models with `gpt-4o` num_retries: 3 timeout: 300 # 30 seconds #cooldown_time: 60 general_settings: store_model_in_db: true store_prompts_in_spend_logs: true # <--- 这是最关键的一行