About Byesu API
Byesu API (半页酥 API) is an AI API gateway. One API key gives you Claude, GPT, Gemini, Grok, GLM, Kimi and image models through OpenAI-compatible and Anthropic-native endpoints, billed by what you use.
What we do
Most teams end up juggling several model providers: separate accounts, separate keys, separate bills and different request formats. Byesu puts them behind one endpoint. You keep the client you already use, whether that is Claude Code, Codex, Cherry Studio or your own code, and change only the base URL and the key.
Requests go to the upstream service for the model you choose. Every call is metered and itemized in your console, so you can see what each request cost and why.
How we work
Four commitments we hold ourselves to, and where you can check them.
The model you ask for
Models are labelled as what they are. We do not serve a cheaper model under a more expensive name.
Billing you can audit
Each request shows its tokens and charge in your usage logs. Model prices are listed on the models and pricing page.
Measured in public
The status page shows live success rate and latency per route. Weekly benchmarks publish time to first token and output speed, with the raw data.
Your data stays yours
We do not store your prompts or model outputs and never use them for training. Details on the security and data page.
Who uses it
Developers who run coding agents every day, small teams building products on top of large language models, and anyone who wants several model families under one bill. Payment by Alipay and USDT is supported.
Talk to us
For production traffic, volume usage or partnerships, write to business@byesu.com. Everything else is on the contact page.
