DeepSeek · Qwen · GLM · Kimi · Seedance

Chinese AI Models API

Run DeepSeek, Qwen, GLM, Kimi, and Seedance from one OpenAI-compatible API. One flatkey key replaces mainland vendor accounts, separate top-ups, and provider-specific SDK work.

Get your Chinese model API key See live pricing

GPT · Gemini · Claude · DeepSeek · Kimi · Seedance — one key, one invoice · no credit card to start

Live-ready model coverage

DeepSeek V4 Flash / 1M tokens	$0.056	$0.14
GLM 5.2 / 1M tokens	$0.56	$1.40
Seedance 2.5 video	Usage-based	Vendor-only
Qwen, Kimi, Hunyuan, Wan	One key	Separate accounts

* Representative catalog coverage — see live pricing for the current per-model rates and access status.

Call Chinese AI models through /v1

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://router.flatkey.ai/v1",
    api_key="YOUR_FLATKEY_KEY",
)

response = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

curl

curl https://router.flatkey.ai/v1/chat/completions \
  -H "Authorization: Bearer $FLATKEY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

China model coverage

DeepSeek, Qwen, GLM, Kimi, Seedance, Kling, Wan, Hailuo, Vidu, MiniMax, Tencent Hunyuan, Baidu ERNIE, and more behind one catalog.

OpenAI-compatible API

Use the OpenAI SDK you already have. Change base_url, set a flatkey API key, and swap model ids like deepseek-v4-flash or glm-5.2.

No mainland vendor setup

Avoid Chinese phone verification, RMB top-ups, local billing profiles, and separate provider consoles when testing or shipping Chinese AI models.

Text, reasoning, and video

Route chat, coding, reasoning, and video generation through the same account, with unified spend controls and one invoice.

Chinese AI Models API

Run DeepSeek, Qwen, GLM, Kimi, and Seedance from one OpenAI-compatible API. One flatkey key replaces mainland vendor accounts, separate top-ups, and provider-specific SDK work.

Get your Chinese model API key

GPT · Gemini · Claude · DeepSeek · Kimi · Seedance — one key, one invoice · no credit card to start

Which Chinese AI model families can I test?

Start with DeepSeek, Qwen, GLM, Kimi, and Seedance, then compare other China model families such as Kling, Wan, Hailuo, Vidu, MiniMax, Hunyuan, and ERNIE as they appear in the catalog.

Is this API compatible with OpenAI SDKs?

Yes. Keep your OpenAI SDK and point it at the flatkey /v1 base URL. The request shape stays familiar; only the base_url, api_key, and model id change.

Do I need a Chinese phone number, RMB account, or local company?

No. flatkey.ai gives international teams one account, one key, and one billing flow for Chinese AI models without managing each mainland vendor account directly.

Do I have to change my code?

No. flatkey.ai is OpenAI-compatible: keep your existing OpenAI SDK and switch base_url plus api_key. Model ids stay the same.

How is billing handled across models?

One plan covers every model. Usage analytics and a single invoice keep spend bounded before you scale.