Get API access to DeepSeek V3, Qwen3, and Kimi K2 ?the models powering the next wave of AI applications.100% OpenAI-compatible.No Chinese phone number. No hassle. Just build.
Everything you need to know before getting started.
Is this really OpenAI-compatible?
Yes, 100%. We implement the full OpenAI chat completions API. Change your base_url and api_key, and your existing code runs without modification. Streaming, function calling, and vision are all supported.
How do I pay?
We now support automated payments via Creem. Pay with credit/debit card (USD). Instant top-up, no subscription. Visit the admin panel to purchase credits. Questions? Contact k112262211@gmail.com.
Where are your servers located?
Our API gateway is hosted in Hong Kong, China, providing low latency for users across Asia-Pacific, North America, and Europe. Global CDN acceleration is planned for the near future.
Do you store my data?
No. We operate a zero-log policy for prompts and responses. Your data is forwarded directly from your client to the model provider. We only track API usage counts for billing purposes.
What models do you support?
Currently: DeepSeek V3 (deepseek-chat), DeepSeek R1 (deepseek-reasoner), Qwen3-235B (qwen3-235b), Kimi K2 (kimi-k2). Check the admin panel for the most up-to-date model list.
How do I get an API key?
Visit our admin panel, register an account, and generate your API key. New accounts get a small test credit to try the API. No credit card required to start.
Ready to Start Building?
Get your API key in under 3 minutes. Start with free test credits — no credit card needed.