APIGOTO – building an independent LLM API gateway

Disclosure: I work on the APIGOTO team. APIGOTO is independent and not affiliated with OpenAI, Anthropic or Google.

APIGOTO is an LLM API gateway: you point any OpenAI-compatible client at one base_url (https://api.china-aiapi.com/v1) and call Claude, GPT and Gemini models through it. When an upstream call fails, the gateway fails over automatically, capped at 3 retries, and every request gets its own log entry. You can bring your own provider keys (BYOK), and billing is pay-per-use.

- one OpenAI-compatible base_url: https://api.china-aiapi.com/v1 (Claude, GPT and Gemini models)

- automatic failover, capped at 3 retries

- per-request logs

- BYOK (bring your own provider keys)

- pay-per-use

Question: is a fixed cap of 3 retries a sensible default, or would you want it configurable per request?

www.china-aiapi.com → https://www.china-aiapi.com/grokbot