AI Model Picker

API Documentation

You can access the model data and our suggestion algorithms programmatically using our REST API.

Endpoints

  • GET /api/models — Full search and suggestion endpoint.
  • GET /api/general — Shortcut for general use suggestions.
  • GET /api/coding — Shortcut for coding tools suggestions.
  • GET /api/value — Shortcut for best value suggestions.
  • GET /api/smart — Shortcut for smart (intel > speed > price) suggestions.

Authentication

All API requests require authentication using an API key. You can provide the API key in one of three ways:

  • Header: x-api-key: YOUR_API_KEY
  • Header: Authorization: Bearer YOUR_API_KEY
  • Query Parameter: ?apiKey=YOUR_API_KEY

Query Parameters

ParameterTypeDescription
searchstringFilter models by name or provider (case-insensitive).
min_speednumberMinimum required speed in output tokens per second.
max_pricenumberMaximum allowed blended price per 1M tokens.
suggestenumApply weighted ranking based on use case. Options: general, coding, value, smart.
limitnumberMaximum number of models to return (default: 10).
expandedbooleanIf true, returns full model object. If false (default), returns only name, slug, and customScore.

Example Requests

# Full search endpoint curl -X GET "http://localhost:3000/api/models?search=llama&limit=3" \ -H "x-api-key: your_secret_key" # Shortcut endpoint for coding tools curl -X GET "http://localhost:3000/api/coding?limit=3&min_speed=50" \ -H "x-api-key: your_secret_key"

Example Response

{
  "data": [
    {
      "name": "Model Name",
      "slug": "model-slug",
      "customScore": 85.4
    }
  ],
  "total": 3,
  "timestamp": "2026-04-12T12:00:00.000Z"
}