Token Control Plane
Redis as an LLM rate limiter and control plane.
A standalone gateway that puts Redis in front of your LLMs as a rate limiter and control plane.
Every request runs a short gauntlet: a router picks the model, a semantic cache short-circuits repeats, a token bucket enforces per-model and per-tenant budgets, and only survivors reach the model. Each stage is a microsecond Redis lookup.
Turns a runaway token bill into something you can see, shape, and cap from one place.