Radar / AI infrastructure / GPUWorks
GPUWorks: a 40MB open-source platform to manage a thousand GPU servers
ZStack open-sourced GPUWorks under Apache 2.0, a 40MB+ platform that deploys in about 60 seconds and manages GPU health, request latency and inference serving across large GPU fleets from one entry point.
Why it matters
A tiny, fast-deploying control plane for GPU inference fleets: one lightweight binary instead of a heavy Kubernetes-plus-monitoring stack, aimed at teams that need model serving without the platform overhead.
What you could build with it
A small team serving several internal models from a rack of GPU servers could deploy GPUWorks to watch GPU health, route inference requests and diagnose failures from one dashboard, replacing a hand-assembled mix of scripts and monitoring tools.
Does it hold up?
Too early to judge: open-sourced hours ago, so no independent deployments or production reports exist yet beyond the vendor announcement.
Built with GPUWorks
- GPUWorks open source: 1,003 GPU servers, 40 MBarticle · Independent breakdown of the announcement numbers, with the caveat that the 1,003-server test is a vendor claim without published GPU counts or network layout.
- GPUWorks launches 40 MB open-source inference platform for heterogeneous GPU clustersarticle · Coverage noting the 1,000+ launch configurations for nearly 400 models across Ascend, T-Head, Hygon and NVIDIA chips, with an OpenAI-compatible API.
- flitai-io/gpuworksgithub · Official open-source repo for the GPUWorks platform, named in the launch announcement.
Learn more
First spotted on article: source.
More AI infrastructure
colibri: run 744B-parameter MoE models on consumer hardwarePure-C inference engine with zero dependencies that streams MoE experts from NVMe disk, keeping hot experts on GPU…infra · JEV 0.78claude-mem: persistent memory plugin that lets coding agents pick up where the last session left offPlugin that hooks five Claude Code lifecycle events, compresses session activity into a local SQLite + Chroma store…infra · JEV 0.74FreeToken: datacenter-scale MoE serving on desktop GPUsAn edge-native mixture-of-experts serving engine from UC Berkeley researchers that runs 290B+ parameter MoE models on…infra · JEV 0.74TorchTPU: Google's native PyTorch stack for TPUs, now open sourceAnnounced at PyTorch Conference 2026 on October 20: Google open-sourced TorchTPU, its native PyTorch backend for Cloud…infra · JEV 0.66Mesh-LLM: pool GPUs across machines behind one OpenAI-compatible endpointRust project that splits large models across machines into layer stages, fetches model shards by package, and serves…infra · JEV 0.64DeepInfer: DeepSeek's open-source production inference engineDeepSeek open-sourced DeepInfer, the high-performance inference framework powering its production services (155…infra · JEV 0.6
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.