One OpenAI-compatible API for Chinese AI models — video, image, audio and LLMs (Seedance · Wan · Kling · Qwen · GLM · MiniMax · DeepSeek V3.2). No Chinese account required.
Batch, compare and storyboard videos with Wan · Seedance · Kling and more.
▶ Open the web app · ⬇ Windows desktop app · Sign up
| Repo | What it is |
|---|---|
| video-studio | Free Windows desktop + MCP Video Studio (batch / compare / storyboard) |
| video-studio-web | Web UI companion — runs in the browser |
| python-examples | Python + OpenAI SDK quickstart |
| typescript-examples | Node / TypeScript + OpenAI SDK quickstart |
| pt-usage-dashboard | Offline usage & cost dashboard (single HTML file, EN/中文) |
| awesome-chinese-model-apis | Curated list of Chinese model APIs and gateways |
Base URL for OpenAI-compatible SDKs: https://api.powertokens.ai/v1
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["POWERTOKENS_API_KEY"],
base_url="https://api.powertokens.ai/v1",
)
resp = client.chat.completions.create(
model="MiniMax-M3",
messages=[{"role": "user", "content": "Describe Paris in one sentence."}],
)
print(resp.choices[0].message.content)Also documented: MiniMax-M3.1-Flash (1M context; docs OpenAPI — pricing page not listed yet).
Links: Website · Docs · Models · Discord
中文：PowerTokens 一个接口调用国产视频 / 图像 / 大模型，无需国内账号 · 网页版 Video Studio · 文档