Quickstart

Tutujin API is compatible with the OpenAI API. Code that already uses an OpenAI SDK needs two changes: set base_url to https://www.tutujin.com/v1 and replace api_key with a key from your console. The whole thing takes about five minutes.

Step 1: Sign up and add credit

1. Create an account on the sign-up page and open the console.
2. Add credit in the console. Balances are in CNY and you pay for what you use — no monthly fee, no minimum spend.
3. For company purchasing or invoices, contact support on WeChat at JsnonoChat.

Step 2: Create an API key

Create a key on the API Keys page of the console. Each key has its own settings:

Keys start with sk-. Use one key per project and per environment, so you can disable one without touching anything else.

Step 3: Set the key and base URL

Put the key in an environment variable; every example below reads it from there:

export TUTUJIN_API_KEY="sk-your-key"

The base URL depends on the API format you use:

The Anthropic and Google SDKs append /v1/messages and /v1beta/models/... themselves, so give them the site root only.

Step 4: Send your first request

Set model to a model name from the model hub. The examples use gpt-5.4-mini:

curl https://www.tutujin.com/v1/chat/completions \
  -H "Authorization: Bearer $TUTUJIN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4-mini",
    "messages": [
      {"role": "user", "content": "Introduce yourself in one sentence"}
    ]
  }'

import os
from openai import OpenAI

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

resp = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[{"role": "user", "content": "Introduce yourself in one sentence"}],
)
print(resp.choices[0].message.content)

import OpenAI from 'openai'

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

const resp = await client.chat.completions.create({
  model: 'gpt-5.4-mini',
  messages: [{ role: 'user', content: 'Introduce yourself in one sentence' }],
})
console.log(resp.choices[0].message.content)

The response has the same shape as OpenAI's. The answer is in choices[0].message.content and this call's usage is in usage:

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "model": "gpt-5.4-mini",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "..." },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 14, "completion_tokens": 32, "total_tokens": 46 }
}

A 401 means the key is wrong or no longer valid; a 403 usually means the balance ran out or the key's allowlists block the call. See Errors and retries for the full list.

Step 5: Check usage and charges

Every call leaves a record in Logs in the console with the model, group, input and output tokens and the cost. For token-billed models, cost = input tokens × input price + output tokens × output price. The example model's cheapest group currently costs:

Billing details are in Billing and usage.

Next steps

• Authentication and base URL: the three header styles and how to configure each native SDK.
• Choosing a group: the same model costs different amounts and runs on different lines in different groups.
• API reference: parameters and examples for every endpoint.
• OpenAI SDK guide: full Python and Node.js setup, including streaming, timeouts and retries.
• Migrating from the official APIs or another OpenAI-compatible service: a checklist, staged rollout and rollback for moving an existing project.