List Models

Retrieve a list of available models from the AI Gateway

GET /v1/models

Returns a list of models available through the gateway. This endpoint can return models from the control plane or pass through to the configured provider.

Authentication

Requires API key authentication:

x-portkey-api-key: YOUR_API_KEY
# OR
Authorization: Bearer YOUR_API_KEY

Request

Headers

x-portkey-api-key · string

Your Portkey API key (alternative to Authorization header)

Authorization · string

Bearer token authentication

x-portkey-provider · string

Optional provider to list models from (e.g., openai, anthropic)

x-portkey-virtual-key · string

Optional virtual key to use for authentication

Query Parameters

Query parameters are passed through to the provider or control plane.

Response

object · string

The object type, always “list”

data · array

Array of model objects

Show model object

id · string

The model identifier (e.g., “gpt-4o”, “claude-3-5-sonnet-20241022”)

object · string

The object type, always “model”

created · integer

Unix timestamp of when the model was created

owned_by · string

Organization that owns the model

Example

cURL

curl https://localhost:8787/v1/models \
  -H "x-portkey-api-key: YOUR_API_KEY"

Response Example

{
  "object": "list",
  "data": [
    {
      "id": "gpt-4o",
      "object": "model",
      "created": 1687882411,
      "owned_by": "openai"
    },
    {
      "id": "gpt-4o-mini",
      "object": "model",
      "created": 1687882411,
      "owned_by": "openai"
    },
    {
      "id": "claude-3-5-sonnet-20241022",
      "object": "model",
      "created": 1687882411,
      "owned_by": "anthropic"
    }
  ]
}

Behavior

The endpoint behavior depends on configuration:

With Provider Header

When you specify x-portkey-provider, the request is forwarded to that provider’s models endpoint:

curl https://localhost:8787/v1/models \
  -H "x-portkey-provider: openai" \
  -H "Authorization: Bearer sk-..."

This returns models available from OpenAI.

Without Provider Header

When no provider is specified and a control plane URL is configured, the gateway returns models from the Portkey control plane, which aggregates models across all configured providers.

Use Cases

Model Discovery

Query available models to dynamically select the best model for your use case based on capabilities, pricing, or performance.

Provider-Specific Models

List models from a specific provider to understand what’s available before making requests.

Model Validation

Check if a specific model ID is available before attempting to use it.

Related Endpoints

Chat Completions

Use a model for chat completions

Completions

Use a model for completions

Embeddings

Use a model for embeddings