model_list:
  - model_name: deepseek-v4-flash
    litellm_params:
      model: deepseek-ai/DeepSeek-V4-Flash
      api_base: https://api.siliconflow.cn/v1
      api_key: os.environ/SILICONFLOW_API_KEY
      rpm: 1000
      tpm: 200000
  - model_name: agnes-2.0-flash
    litellm_params:
      model: openai/agnes-2.0-flash
      api_base: https://ai.qihuapi.com/v1
      api_key: os.environ/AGNES_API_KEY
      rpm: 500
  - model_name: mimo-v2.5
    litellm_params:
      model: openai/mimo-v2.5
      api_base: https://token-plan-cn.xiaomimimo.com/v1
      api_key: os.environ/MIMO_API_KEY
      rpm: 500
  - model_name: qwen3-vl-32b
    litellm_params:
      model: openai/Qwen3-VL-32B-Instruct
      api_base: https://api.siliconflow.cn/v1
      api_key: os.environ/SILICONFLOW_API_KEY
      rpm: 300

litellm_settings:
  drop_params: true
  request_timeout: 120

general_settings:
  master_key: os.environ/LITELLM_MASTER_KEY

router_settings:
  routing_strategy: "usage-based-routing-v2"
  allowed_fails: 2
  num_retries: 2
  fallbacks:
    - deepseek-v4-flash: ["agnes-2.0-flash"]
    - agnes-2.0-flash: ["deepseek-v4-flash"]
    - mimo-v2.5: ["deepseek-v4-flash"]