Models

List Models

curl "https://cavos.org/api/models?search=gpt-4&limit=10"

Query Parameters

Parameter	Type	Description
search	string	Full-text search across model names and descriptions
provider	string	Filter by provider slug
modality	string	Filter by modality (text, image, etc.)
contextLength	number	Minimum context length
page	number	Page number (default: 1)
limit	number	Results per page (default: 20)

Response

{
  "models": [
    {
      "id": "model-uuid",
      "modelId": "openai/gpt-4o",
      "name": "GPT-4o",
      "slug": "openai-gpt-4o",
      "description": "Most capable GPT-4 model",
      "contextLength": 128000,
      "promptPricing": 2.5,
      "completionPricing": 10.0,
      "imagePricing": null,
      "provider": {
        "name": "OpenAI",
        "slug": "openai"
      },
      "architecture": {
        "modality": "text+image->text",
        "tokenizer": "GPT",
        "instructType": "none"
      }
    }
  ],
  "total": 250,
  "page": 1,
  "limit": 20
}

List Providers

curl https://cavos.org/api/providers

Returns all available model providers with their slugs and model counts.

Pricing

Model pricing is per million tokens in USD:

Field	Description
promptPricing	Cost per 1M input tokens
completionPricing	Cost per 1M output tokens
imagePricing	Cost per image (for image models)

Cost Calculation

cost = (promptTokens × promptPricing / 1,000,000)
     + (completionTokens × completionPricing / 1,000,000)

Model Sync

Models and pricing are synced from OpenRouter periodically. New models become available automatically.