⬢github Go · 109 ★ +7 since we first saw it · pushed 1 d ago · Apache-2.0
flitai-io/gpuworks
A ~45 MB MaaS platform: one cluster across public and private clouds, on any mix of accelerators.
GPUWorks is a lightweight Go-based GPU cluster manager and Model-as-a-Service platform. It pools your own servers, rented cloud instances, and containers—including mixed NVIDIA, AMD, Ascend, Hygon, and Apple Silicon hardware—into one cluster, then deploys models with vLLM, SGLang, or other engines behind a single OpenAI-compatible endpoint. The whole control plane is a ~40 MB static binary with an embedded web UI and SQLite storage, no Kubernetes required.
Why now: Recently launched via a Show HN post emphasizing its tiny 40 MB footprint and no-Kubernetes approach as a contrast to heavier GPU orchestration stacks.
Who it is for: Teams running LLM inference on heterogeneous GPU hardware across cloud and on-prem machines who want simple orchestration without Kubernetes.
Stars over our 14 snapshots: 102 to 109, since 1 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/flitai-io/gpuworks