About

genmux API is a unified HTTP API for image-generation and chat (text) models. Instead of integrating each model provider separately, you integrate one endpoint, hold one credit balance, and switch models by changing a single request field.

The platform is built around one guarantee: exactly-once billing. A credit is captured only when a job succeeds — per image for image models, per token for text models; failed and cancelled jobs are refunded automatically. Under the hood every generation is an asynchronous job with idempotency keys, webhooks and per-key rate limits — the boring, reliable plumbing you would otherwise build yourself.

What we run

A queue-backed generation pipeline with automatic failover between upstream providers, OpenAI-compatible chat completions with streaming and tool calling, per-model price transparency, and a dashboard with usage, ledger and API-key management. Prompts and chat messages are sent to the model provider’s gateway only to fulfil your request and are never used to train models. Card payments and USDT deposits are both supported.

Contact

Questions, volume pricing, or a model you want us to add: support@genmux.tech.