[rustybin]
A high performance, Rust-based httpbin alternative that mimics backend and LLM responses for API gateway load testing and AI mocking.
- Status
- In development
- Stack
- Rust · HTTP · LLM mocks
What it is
An HTTP server written in Rust that answers the usual httpbin endpoints and adds a set of mock responses shaped like real backends and LLM providers. Point an API gateway at it and you can load test the gateway without an upstream that costs money, rate limits you, or falls over first.
Why it exists
Load testing a gateway against a real upstream measures the upstream as much as the gateway. The same goes for AI gateways: a request to a hosted model takes seconds and is billed per token, which makes a ten minute soak test expensive and slow. A small, fast, predictable upstream removes the variable.
What it does
- Serves the httpbin endpoints people actually use: status codes, delays, headers, redirects, JSON echo.
- Mimics chat completion and embedding responses in the formats the main providers use, with configurable latency and token streaming.
- Runs as a single static binary with no runtime dependencies.
Status
In development. The endpoint set and the configuration format are still changing. The name is a placeholder. Documentation lives at docs.rustybin.dev.