Models endpoint

GET /v1/models returns the models the current key can call: the ones its group offers that also pass the key's model allowlist. The default response is in OpenAI format; with Claude or Gemini auth headers you get the format their official SDKs parse.

Request example

curl https://open.tutujin.com/v1/models \
  -H "Authorization: Bearer $TUTUJIN_API_KEY"

import os
from openai import OpenAI

client = OpenAI(base_url="https://open.tutujin.com/v1", api_key=os.environ["TUTUJIN_API_KEY"])

for m in client.models.list():
    print(m.id)

import OpenAI from 'openai'

const client = new OpenAI({ baseURL: 'https://open.tutujin.com/v1', apiKey: process.env.TUTUJIN_API_KEY })

for await (const m of client.models.list()) {
  console.log(m.id)
}

Response (OpenAI format)

The structure, with placeholder values:

{
  "success": true,
  "object": "list",
  "data": [
    {
      "id": "gpt-5.4-mini",
      "object": "model",
      "created": 1700000000,
      "owned_by": "openai",
      "supported_endpoint_types": ["openai"]
    }
  ]
}

The key decides what is listed: a key in another group may return different models. For every model and its price in each group, see Models and pricing.

Claude and Gemini formats

With the x-api-key and anthropic-version headers, the response uses Anthropic's model-list format, which the Anthropic SDK's client.models.list() parses directly:

curl https://open.tutujin.com/v1/models \
  -H "x-api-key: $TUTUJIN_API_KEY" \
  -H "anthropic-version: 2023-06-01"

{
  "data": [
    {
      "id": "claude-sonnet-5-5",
      "type": "model",
      "display_name": "claude-sonnet-5-5",
      "created_at": "2023-11-14T22:13:20Z"
    }
  ],
  "first_id": "...",
  "last_id": "...",
  "has_more": false
}

With an x-goog-api-key header, or on GET /v1beta/models, you get Gemini's format:

curl https://open.tutujin.com/v1beta/models \
  -H "x-goog-api-key: $TUTUJIN_API_KEY"

{
  "models": [
    {
      "name": "gemini-3.8-flash",
      "displayName": "gemini-3.8-flash"
    }
  ],
  "nextPageToken": null
}

In the Gemini format only name and displayName are filled in; the other fields are empty, and name is the bare model name without a models/ prefix.

Notes

• The list only contains models this key can use right now. A key with a model allowlist only returns the allowlisted models.
• GET /v1/models/{model} looks up a single model, but it only recognizes some common model names and returns an error object with model_not_found for the rest. To check whether you can call a model, rely on the list.
• The list changes rarely; fetch it when your service starts and cache it rather than querying before every request.