Use DeepSeek, Qwen, GLM and Kimi through one OpenAI-compatible API. Unified tokens, billing, rate limits and routing for developers.
One token for mainstream LLMs. View the full model list.
No subscription lock-in. Recharge what you need, spend by token, and keep your balance valid.
Register, create a token, and switch your client base URL.
Email registration. Recharge by the available payment methods in your region.
Set quota, expiration and allowed model groups from the dashboard.
curl https://we2ai.ai/v1/chat/completions \
-H "Authorization: Bearer sk-xxxx" \
-d '{"model":"deepseek-chat",
"messages":[{"role":"user",
"content":"hello"}]}'You get one token, one bill, one endpoint and one set of logs instead of registering and funding every provider separately.
We only keep request metadata such as time, model and token usage for billing and troubleshooting. Prompt and response bodies are not retained by default.
Routes can be configured with multiple upstream channels. If one channel fails, traffic can be shifted to another compatible channel.