GET /v1/models — List Available Models

List all models available through CoreValue Gateway, with auth-required filtering.

How It Works

The GET /v1/models endpoint returns an OpenAI-compatible list of models. It requires authentication (unlike some other gateways).

Usage

curl https://gateway.corevalue.dev/v1/models \
  -H "Authorization: Bearer sk-cova-..."

Response:

{
  "object": "list",
  "data": [
    { "id": "gpt-4o", "object": "model", "created": 1699999999, "owned_by": "openai" },
    { "id": "claude-3-5-sonnet", "object": "model", "created": 1699999999, "owned_by": "anthropic" }
  ]
}

Auth Required

This endpoint requires authentication (Authorization: Bearer or api-key header). Requests without auth return 401.

requireExplicitRouting Filter

Models where all endpoints have requireExplicitRouting === true are excluded from the list. This hides internal/routing-only models from clients. The filter is server-side — it cannot be bypassed by client query parameters.

Caching

The model list is cached in-memory for 60 seconds. New models may take up to 60 seconds to appear.

Use Case: Dynamic Model Selection

const response = await fetch("https://gateway.corevalue.dev/v1/models", {
  headers: { "Authorization": "Bearer sk-cova-..." },
});
const { data: models } = await response.json();
const modelIds = models.map(m => m.id);