Amux AI

List Models

Last updated August 26, 2026

Returns the currently callable models in Gemini's response shape, with a few practical extension fields.

Returns the currently callable models in Gemini's response shape. The Gemini SDK's model-listing call can target this endpoint directly.

GEThttps://gateway.amux.ai/v1beta/models

Authorization

header
AuthorizationstringRequired

The Amux key created in the console. This is the header the Gemini SDKs send natively; Authorization: Bearer and ?key= on the URL are also accepted.

Request

pageSizeinteger

Pagination size on Gemini's official list method. Not supported here — this endpoint always returns the full set of currently callable models.

pageTokenstring

Pagination cursor on Gemini's official list method. Not supported here — this endpoint does not paginate, so there is no next page to fetch.

Response

200response

All currently callable models in Gemini's shape. name carries the models/ prefix so it can be concatenated straight into the next request's URL; this endpoint does not paginate, so it also returns no nextPageToken.

401response

The key is missing or invalid.

What a key changes

This endpoint does not require authentication — hitting it right after you change base_url is the first step of any integration.

When you send a valid API key, you get back the models that key can actually call: its model allowlist, source scope and pinned providers all apply here. Prices also reflect your account's discounts.

An invalid, expired or disabled key is never an error — it is treated as unauthenticated and you get the full list. So this endpoint is not an access check; the real decision happens when you make the call.

The response uses Gemini's shape

{ "name": "models/anthropic/claude-opus-5", "baseModelId": "anthropic/claude-opus-5" }

name carries the models/ prefix so it can be concatenated straight into the next request's URL ({name}:generateContent). For the model ID on its own, use baseModelId.

This endpoint returns the full list in one response, so it does not return the official list method's nextPageToken.

Main differences from the other two

OpenAI / AnthropicGemini
Top-level arraydatamodels
Field stylesnake_casecamelCase
Context windowcontext_length / max_input_tokensinputTokenLimit
Max output per callmax_output / max_tokensoutputTokenLimit

supportedGenerationMethods is fixed at ["generateContent", "streamGenerateContent"]. Whether a call streams is decided per request, not by the model.

There is no description field. Model descriptions are multilingual; to read them in a given language, see the model pages.

Differences from Gemini's official list method

pageSize and pageToken are not supported here. This endpoint always returns the full set of currently callable models, so there is no pagination cursor to continue with.

Extra fields

thinking, aliases, pricing, and providers are not official Gemini fields. Gemini SDKs ignore them; over plain HTTP you can read them as extra metadata.

  • thinking: boolean convenience flag indicating whether the model currently exposes reasoning capability.
  • aliases, pricing, and providers: same meaning as on the OpenAI-shape page.

Errors

Errors come back in Gemini's error shape. See Errors and retries.

cURL
curl https://gateway.amux.ai/v1beta/models \
  -H "x-goog-api-key: $AMUX_API_KEY"
{
  "models": [
    {
      "name": "models/google/gemini-3-pro",
      "baseModelId": "google/gemini-3-pro",
      "displayName": "Gemini 3 Pro",
      "inputTokenLimit": 1048576,
      "outputTokenLimit": 65536,
      "supportedGenerationMethods": [
        "generateContent"
      ],
      "thinking": false,
      "aliases": [
        "<string>"
      ],
      "pricing": {},
      "providers": [
        {}
      ]
    }
  ]
}