Serve is a prototype console. Works today: a browser playground that returns a simulated action. 4-bit quantisation, Prometheus metrics, request batching, and a Docker single-command deploy are not built.
Vision-Language-Action inference endpoint and REST containerOHH-36
Future
Vision: not built
Robot AI inference, as an API.
Robot AI inference, as an API. Deploy a Vision-Language-Action model behind a REST endpoint in one command — pluggable model backends, batching, and Prometheus metrics built in.
Vision highlights
One-command model deploy
REST predict API
Pluggable backends (OpenVLA, SmolVLA, ACT…)
4-bit quantization option
Prometheus metrics built in
Planned specifications
API
REST — /health, /load_model, /predict
Backends
OpenVLA, SmolVLA, ACT, diffusion, custom
Quantization
Optional 4-bit
Hardware
Your GPU (desktop, server or cloud) — we help you size it