LLM provider targets call language model APIs directly. These are used both as evaluation targets and as judge targets for scoring.
endpoint: ${{ AZURE_OPENAI_ENDPOINT }}
api_key: ${{ AZURE_OPENAI_API_KEY }}
model: ${{ AZURE_DEPLOYMENT_NAME }}
| Field | Required | Description |
|---|
endpoint | Yes | Azure OpenAI endpoint URL |
api_key | Yes | API key |
model | Yes | Deployment name |
api_key: ${{ ANTHROPIC_API_KEY }}
model: claude-sonnet-4-20250514
| Field | Required | Description |
|---|
api_key | Yes | Anthropic API key |
model | Yes | Model identifier |
api_key: ${{ GEMINI_API_KEY }}
| Field | Required | Description |
|---|
api_key | Yes | Google AI API key |
model | Yes | Model identifier |