API Documentation

OpenAI-compatible API gateway to top Chinese AI models — DeepSeek, Qwen, GLM, MiniMax. Use your existing SDK, just change the base URL.

Quick Start

1. Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-YOUR_API_KEY",
    base_url="https://api.tunanapi.com/v1"
)

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

2. cURL

curl https://api.tunanapi.com/v1/chat/completions \
  -H "Authorization: Bearer sk-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

3. Node.js

import OpenAI from 'openai';

const client = new OpenAI({
  apiKey: 'sk-YOUR_API_KEY',
  baseURL: 'https://api.tunanapi.com/v1'
});

const response = await client.chat.completions.create({
  model: 'deepseek-chat',
  messages: [{ role: 'user', content: 'Hello!' }]
});

console.log(response.choices[0].message.content);

Authentication

All requests require an API key in the Authorization header:

Authorization: Bearer sk-YOUR_API_KEY

Get your key at tunanapi.com → Sign up with email → Copy API key from dashboard.

Base URL & SDK Configuration

Models & Pricing

All prices per 1M tokens. Input = prompt tokens, Output = completion tokens. Billed per token, no minimums.

Flagship Models

Fast & Affordable

Specialized Models

Chat Completions

{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "created": 1718900000,
  "model": "deepseek-chat",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 8,
    "completion_tokens": 9,
    "total_tokens": 17
  }
}

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "system", "content": "You are a helpful coding assistant."},
        {"role": "user", "content": "Write a Python function to reverse a linked list"},
    ]
)

response = client.chat.completions.create(
    model="deepseek-chat",
    response_format={"type": "json_object"},
    messages=[{"role": "user", "content": "Return a JSON object with name, age, city"}]
)

List Models

curl https://api.tunanapi.com/v1/models \
  -H "Authorization: Bearer sk-YOUR_API_KEY"

{
  "object": "list",
  "data": [
    {"id": "deepseek-chat", "object": "model", "owned_by": "deepseek"},
    {"id": "qwen3.7-max", "object": "model", "owned_by": "qwen"},
    ...
  ]
}

Embeddings

response = client.embeddings.create(
    model="text-embedding-v3",
    input="Hello world"
)

print(response.data[0].embedding[:5])  # [0.0023, -0.0094, ...]

Vision

response = client.chat.completions.create(
    model="qwen-vl-max",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "What's in this image?"},
            {"type": "image_url", "image_url": {"url": "https://example.com/photo.jpg"}}
        ]
    }]
)

response = client.chat.completions.create(
    model="qwen-vl-max",
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "Describe this image"},
            {"type": "image_url", "image_url": {"url": "data:image/jpeg;base64,/9j/4AAQ..."}}
        ]
    }]
)

Streaming

Set stream: true to receive partial results as Server-Sent Events (SSE). Essential for chat interfaces and long responses.

stream = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Write a poem about the sea"}],
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")

Raw SSE format (cURL)

curl https://api.tunanapi.com/v1/chat/completions \
  -H "Authorization: Bearer sk-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"deepseek-chat","messages":[{"role":"user","content":"Hi"}],"stream":true}'

# Each chunk:
data: {"id":"chatcmpl-...","choices":[{"delta":{"content":"Hello"}}]}
data: [DONE]

Fallback & Routing

How it works

When you call a model like deepseek-chat, TunanAPI routes your request to the primary provider. If that provider returns a 5xx error or times out, the system automatically retries on a configured backup channel. You get:

• Higher uptime — one provider going down doesn't break your app
• Zero code changes — just call the model as usual
• Transparent billing — you're billed by the model that actually served the request

Rate Limits

Rate limit headers

Every response includes headers to help you track your usage:

Error Codes

Error response format

{
  "error": {
    "message": "Insufficient quota. Please top up your account.",
    "type": "insufficient_quota",
    "code": 402
  }
}

Retry strategy

Billing & Quota

How billing works

You're billed per token, per the pricing table above. There are no subscriptions or minimums — pay only for what you use.

Check your balance

# Using API
curl https://api.tunanapi.com/api/user/self \
  -H "Authorization: Bearer sk-YOUR_API_KEY"

# Response includes:
{"data": {"quota": 500000, "used_quota": 12345, ...}}

Quota is in units where 1 unit = $0.002 / 1K tokens (1,000,000 units = $2).

Top up

Visit tunanapi.com dashboard → Top Up. We accept credit cards and PayPal.

Pricing Tiers

Integrations

LangChain

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="deepseek-chat",
    api_key="sk-YOUR_API_KEY",
    base_url="https://api.tunanapi.com/v1"
)

response = llm.invoke("Explain quantum computing in one paragraph")

LlamaIndex

from llama_index.llms.openai import OpenAI as LlamaOpenAI

llm = LlamaOpenAI(
    model="deepseek-chat",
    api_key="sk-YOUR_API_KEY",
    api_base="https://api.tunanapi.com/v1"
)

CrewAI

from crewai import Agent, Task, Crew
import os

os.environ["OPENAI_API_KEY"] = "sk-YOUR_API_KEY"
os.environ["OPENAI_API_BASE"] = "https://api.tunanapi.com/v1"

agent = Agent(role="Researcher", goal="Find insights", backstory="Expert researcher")

AutoGen

import autogen

config_list = [{
    "model": "deepseek-chat",
    "api_key": "sk-YOUR_API_KEY",
    "base_url": "https://api.tunanapi.com/v1"
}]

assistant = autogen.AssistantAgent("assistant", llm_config={"config_list": config_list})

Migration from OpenAI

# Before (OpenAI)
client = OpenAI(api_key="sk-openai-...")  # uses api.openai.com

# After (TunanAPI)
client = OpenAI(
    api_key="sk-YOUR_TUNANAPI_KEY",
    base_url="https://api.tunanapi.com/v1"
)

# Everything else is identical!

Model mapping

Changelog

2026-06-14

• Launched comprehensive API documentation
• Added Fallback & Routing documentation
• Added Vision API and Embeddings documentation
• Added integration guides (LangChain, LlamaIndex, CrewAI, AutoGen)

2026-06-09

• Pricing V2.1上线 — 混合毛利率策略，8 models online
• MiniMax M3 (1M context) added

2026-06-07

• I Ching Oracle product launch at oracle.tunanapi.com
• Terms of Service & Privacy Policy published