Model APIs for business

Model APIs for companies building AI products

724 models behind one OpenAI-compatible endpoint. The price of every model in every group is public. You pay for what you use, with no monthly fee.

724 models
18 groups
OpenAI-compatible
Pay as you go

Popular model prices

Prices updated Nov 12, 2026, 8:17 PM

Model	Cheapest group	Input	Output
gpt-6.1-sol StandardMixed sources	StandardMixed sources	¥4	¥20
claude-haiku-5-5 Claude PlusMixed sources	Claude PlusMixed sources	¥0.4	¥2
gemini-3.8-flash StandardMixed sources	StandardMixed sources	¥1.5	¥7.5
deepseek-v4.1-flash China Models	China Models	¥2.8	¥11.2
qwen3.5-plus China Models	China Models	¥1.12	¥6.72
kimi-k3.1-pro China Models	China Models	¥18.2	¥95.19
gpt-5.5 StandardMixed sources	StandardMixed sources	¥10	¥80
claude-sonnet-5-5 Claude LiteMixed sources	Claude LiteMixed sources	¥3.2	¥16

¥ per 1M tokens, cheapest group shown for each modelAll 724 models

Integration

One endpoint for every model

If your code already uses an OpenAI SDK, point base_url at us and use one of our API keys. Native Claude and Gemini formats work too.

Available endpoints

Chat Completions: /v1/chat/completions
Responses: /v1/responses
Claude native: /v1/messages
Gemini native: /v1beta/models/{model}:generateContent
Image generation / editing: /v1/images/generations · /v1/images/edits
Embeddings: /v1/embeddings
Rerank: /v1/rerank
Speech / transcription: /v1/audio/speech · /v1/audio/transcriptions

Create API keys in the console. The samples read the key from the TUTUJIN_API_KEY environment variable; never ship it in front-end code.

Read the API docs

Base URLhttps://open.tutujin.com/v1

import os
from openai import OpenAI

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

resp = client.chat.completions.create(
    model="gpt-6.1-sol",
    messages=[{"role": "user", "content": "Introduce yourself in one sentence."}],
)
print(resp.choices[0].message.content)

Why Tutujin

Things you can check yourself before you integrate

Prices, route sources, usage and account security are all visible on the site or in the console.

Transparent pricing

Every model's price in every group is public, with input, output and cache priced separately. Export a quote for finance or procurement in one click.

Example gpt-6.1-sol/ 1M tokens

Group	In	Out
StandardMixed sourcesCheapest	In¥4	Out¥20
AzureCloud	In¥4	Out¥20
Azure PlusCloud	In¥6	Out¥30
OpenAI DirectOfficial	In¥9	Out¥45
OpenAI Direct PlusOfficial	In¥12	Out¥60
Official · EnterpriseOfficial	In¥21.9	Out¥109.5

See all prices and export a quote

Every group labelled by source

One model often has several routes. Each group states where it comes from, and groups that include non-official sources say so next to their name.

Official7 groupsRelays the model vendor's own official API
Cloud2 groupsOfficial model service from a cloud provider, such as Azure OpenAI
Mixed sources6 groupsMixed sources including non-official ones, usually cheaper
Lowest priceThe cheapest group for a given model

If the source matters for your use case, pick an Official or Cloud group. If price matters more and non-official sources are acceptable, consider a Mixed group.

Read the group notes

Itemized usage and per-key caps

Every call shows up in the console under Usage with its token count and cost. Each API key can also have its own:

Spending capThe key stops working once its quota is used up
Expiry dateExpires automatically, handy for short projects
Model allowlistOnly the models you list can be called
IP allowlistOnly requests from listed IPs are accepted
GroupEach key uses the group you assign

Account security

Sign-in supports two-factor authentication and passkeys, and sign-ins and security changes are recorded in an audit log.

Two-factor (TOTP)Sign-in also asks for a code from your authenticator app
PasskeysSign in with your device's fingerprint, face or security key
Session managementSee your signed-in sessions and revoke them
Audit logSign-ins and security changes are recorded
Personal access tokensScoped tokens for scripts and automation
Encrypted password sign-inHTTPS throughout; passwords are encrypted before they are sent

Security & compliance

Procurement and invoices

Top up online (as shown in the console) or buy redemption codes offline through support.

We can issue invoices (add our WeChat support and send your order number). For China domestic fapiao, contact us to discuss.

Enterprise

Use cases

Common ways companies use the API

Open a use case to see how it is put together.

All solutions

Three steps

From sign-up to your first call

Your existing calling code barely changes.

1
Sign up and top up
Create an account and add credit. You are charged for actual usage, with no monthly fee.
Sign up free
2
Create an API key and pick a group
Create a key in the console and choose a group. Add a spending cap and model or IP allowlists if you need them.
Open the console
3
Point base_url at us
Use your Tutujin API key and the model names listed on the models page.
The base_url change
```
- base_url = "https://api.openai.com/v1"+ base_url = "https://open.tutujin.com/v1"
```
Quickstart

FAQ

Questions people ask before integrating

Can't find your answer? Ask our support team.

What is Tutujin API?

We supply model APIs to companies that build AI products: 724 models including GPT, Claude, Gemini, DeepSeek and Qwen, all called through one OpenAI-compatible endpoint. We don't build end-user AI apps ourselves.

How is it billed? Is there a monthly fee or minimum spend?

You pay as you go. There is no monthly fee and no minimum spend commitment, and ¥1 buys ¥1 of credit. Chat models are billed per token, with input, output and cache reads/writes priced separately; per-request models such as image generation are billed per request.

Why does one model have several prices?

A model usually sits in several groups, and each group is a different route: official relay, cloud provider, or mixed sources that include non-official ones. Different sources mean different prices. Every group is labelled with its source, and you choose one when you create an API key.

Which API formats are supported?

The OpenAI-compatible format (Chat Completions and Responses), Claude's native /v1/messages and Gemini's native generateContent. Image generation and editing, embeddings, rerank, speech and transcription are available too.

Where can I see what each call cost?

The Usage page in the console lists every call with its model, token count and cost.

Can I limit what an API key can spend or call?

Yes. Each key can have its own spending cap, expiry date, model allowlist and IP allowlist, and it uses the group you assign.

Do you store the content of my requests?

Usage logs record call details only: time, model, group, key name, token counts, cost and latency. Request and response bodies are not written to usage logs. To generate a response, your request is forwarded to the model provider behind the group you chose. See the Security & compliance page for details.

How do I top up? Can companies buy in bulk?

Top up online (as shown in the console) or buy redemption codes offline through support. For bulk purchases, add our support on WeChat.

Can you issue invoices?

Yes, we can issue invoices: add our WeChat support and send your order number. For China domestic fapiao, contact us to discuss.

Sign up and start calling

For procurement, invoices or help with integration, add our support on WeChat.

Pay as you go, no monthly fee or minimum spend
Invoices available

WeChatJsnonoChat

Scan the code or search the WeChat ID to add us