2026-05-27 · API Guide · Updated 2026-07-05 · 约 5 分钟阅读 · Editor: MiMo AI Community

Xiaomi MiMo API Integration Guide

Quick Start

The MiMo API is OpenAI-compatible. You can integrate it with a simple curl command or the Python OpenAI SDK:

curl https://api.xiaomimimo.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $MIMO_API_KEY" \
  -d '{
    "model": "mimo-v2.5-pro",
    "messages": [{"role": "user", "content": "Hello, what can you do?"}],
    "temperature": 0.7,
    "max_tokens": 1024
  }'

1. Authentication

To access the MiMo API, you need an API key from the Xiaomi MiMo Platform:

1. Sign up at platform.xiaomimimo.com
2. Navigate to API Keys → Create New Key
3. Use the key in the Authorization: Bearer <key> header
4. New accounts receive $2 in free trial credits (no credit card required)

2. Available Models

3. Python Integration

Using the OpenAI Python SDK (v1.0+):

from openai import OpenAI

client = OpenAI(
    api_key="your-mimo-api-key",
    base_url="https://api.xiaomimimo.com/v1"
)

response = client.chat.completions.create(
    model="mimo-v2.5-pro",
    messages=[
        {"role": "system", "content": "You are MiMo, a helpful AI assistant."},
        {"role": "user", "content": "Explain MoE architecture in 3 sentences."}
    ],
    temperature=0.7,
    max_tokens=512
)

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

4. Token Plan Subscription

The MiMo Token Plan is a subscription bundle designed for AI coding scenarios. It offers high-value credit packages for using MiMo flagship models in popular coding tools:

• Coding-focused: Optimized for developers using MiMo Code, Cursor, Cline, Zed
• Auto-renewal: Supports automatic renewal (enabled from April 2026)
• 5-8× more credits: Recent upgrade increased credits quota without price increase
• Global availability: Available worldwide since May 2026

5. Migration from OpenAI

Migrating from OpenAI to MiMo API requires minimal code changes:

Just change the base_url and model name — the API is fully compatible!

6. Rate Limits & Best Practices

• Rate limits: Vary by plan tier; Token Plan subscribers get higher limits
• Retry logic: Implement exponential backoff for 429/503 responses
• Streaming: Use stream: true for real-time responses
• Tool calling: MiMo-V2.5-Pro supports native function calling for agent workflows

7. API Endpoints