Skip to main content

ai.yml

Design

Each provider represents an API endpoint (base_url + api_key) that can host multiple models.

Models are referenced as provider_id/model_id, e.g.:

  • openai/gpt-4o
  • deepseek/deepseek-chat
  • ollama/qwen2.5:7b

Refer to the full example above for the YAML structure.

Field Reference

providers[]

FieldTypeDescription
idstringUnique provider ID
namestringDisplay name
typestringopenai / openai-compatible / deepseek / ollama
base_urlstringAPI endpoint URL
api_keystringAPI key (supports ${ENV} placeholder)
modelsarrayList of models under this endpoint

models[]

FieldTypeDefaultDescription
idstringModel name passed to the API
namestringDisplay name in the UI
max_tokensintMax output tokens
temperaturefloat0.7Sampling temperature
thinking_enabledboolfalseDeepSeek only: Enable thinking mode
reasoning_effortstringhighDeepSeek only: Reasoning intensity (low/medium/high)

default_model

The default selected model, in provider_id/model_id format.

DeepSeek Thinking Mode

When type: "deepseek" and thinking_enabled: true, the system automatically enables tRPC-Agent-Go's BuiltinPlanner:

  1. Injects thinking parameters before each request (via reasoning_effort)
  2. Extracts reasoning_content from responses
  3. Preserves chain-of-thought context across multi-turn conversations

This ensures DeepSeek's reasoning_content is correctly managed throughout the entire conversation.