Can I keep the OpenAI SDK?
Yes. In most server-side SDK code, keep the OpenAI-style client and change the base URL to https://www.tken.shop/v1. Use your TKEN API key as the bearer token.
OpenAI-compatible API gateway
One base URL for multiple AI model families
TKEN gives developers an OpenAI-compatible gateway for testing GPT, Claude, DeepSeek, Qwen, Kimi, MiniMax, GLM and other model families from one API key and one integration pattern.
Gateway endpoint
https://www.tken.shop/v1
curl https://www.tken.shop/v1/chat/completions \
-H "Authorization: Bearer $TKEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "MiniMax-M2.7",
"messages": [
{"role": "user", "content": "Create a launch checklist."}
]
}'OpenAI-compatible chat completions
One API key for model comparison
Designed for developer tooling
Built for migration and model testing
Use this page when evaluating a custom OpenAI-compatible provider, testing Chinese and global model families, or routing traffic from tools that already expose an API base URL setting.
Base URL
https://www.tken.shop/v1
Primary flow
Chat completions
Best fit
SDKs, web UIs, agent tooling
Disclosure
Independent third-party gateway
OpenAI-compatible gateway FAQ
Is TKEN an official endpoint for the model providers?
No. TKEN is an independent third-party API gateway. It should not be described as an official OpenAI, Anthropic, Google, DeepSeek, Qwen, Kimi, MiniMax or GLM endpoint.
How should teams check availability and price?
Check the live pricing page before production use. Model availability, names, routing and prices can change as upstream provider access changes.
Test one OpenAI-compatible endpoint
Start with a small request, verify the model response, then route the same integration through your preferred developer tool.