# Project-HAMi/HAMi > HAMi is a Kubernetes middleware for virtualizing and sharing GPUs and other AI accelerators: it lets pods request a fraction of a device by memory or cores and schedules them with device-aware policies. A CNCF project aimed at platform teams running shared AI clusters. - Magnitude: 2.3 out of 10 — Steady - Stars: 4,678 total · +2 stars measured 2026-10-11, 11:04–12:31 UTC, ≈ 10 by evening - Star trust: star growth looks organic - Category: DevOps & infrastructure · Language: Go · License: Apache-2.0 · Created: 2021-09-14 · Last push: 2026-10-10 - GitHub: https://github.com/Project-HAMi/HAMi · Homepage: https://project-hami.io · Page: https://gitnova.dev/en/r/Project-HAMi/HAMi ## Useful for - Allocate a fraction of a GPU to an inference pod so several models share one card - Configure binpack scheduling to consolidate AI workloads onto fewer nodes - Collect accelerator usage metrics via /metrics and Grafana dashboards ## Why it’s here - Star-counter measurements on 2026-10-11 (UTC), 11:04–12:31: 4676 → 4678 stars (+2). This is the change over that interval. - Estimated end-of-day forecast: about +10 stars, using observed gains and the previous day. - Over the last two days the pace is 5× that of the previous week and a half. - Elevated gains were measured on the last 2 completed UTC days in a row. Continued growth is not guaranteed. - GitHub Trending Go today: #8, +35 stars. ## Star trust Star growth looks organic. Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer. ## Numbers - Forks: 853 - Issues and pull requests: 3,191 - Watchers: 27 - Average over the last week: 12 per day - Usual pace: 5 per day - Stars in the last hour (measured): 1 - Latest release: v2.10.0 (2026-08-21) ## Stars per day, last 30 days (oldest → newest, today is partial) 2026-09-12 … 2026-10-11: 2, 8, 10, 11, 11, 14, 8, 6, 8, 8, 5, 5, 1, 5, 4, 7, 4, 3, 7, 4, 2, 2, 2, 6, 3, 5, 7, 20, 31, 2 ## Spotted in now - GitHub Trending Go today: #8, +35 stars ## More in this category 1. **t4t5/omdrop-owl** — 8.3 · Breakout · DevOps & infrastructure · Shell · +421 stars measured 2026-10-11, 00:38–12:27 UTC, ≈ 842 by evening Radio backend for omdrop that brings AirDrop to non-Apple Wi-Fi cards. It runs AWDL in userspace via OWL on a monitor interface, targeting the MediaTek MT7925 while Wi-Fi stays connected. Full card: https://gitnova.dev/en/r/t4t5/omdrop-owl.md 2. **Nongfsq/ftop** — 4.4 · Early signal · DevOps & infrastructure · Swift · +26 stars measured 2026-10-11, 00:39–12:28 UTC, ≈ 60 by evening A small system monitor for Apple Silicon Macs: one floating glass panel showing live CPU cores, memory, GPU, network, power, and processes. Full card: https://gitnova.dev/en/r/Nongfsq/ftop.md 3. **ToiCF/CF-Workers-TGProxy** — 4.3 · Cooling · DevOps & infrastructure · +0 stars measured 2026-10-11, 00:39–12:28 UTC, ≈ 25 by evening A single-file Cloudflare Workers relay that accepts Telegram Web/Desktop WebSocket traffic and forwards it to Telegram Web K's official KWS endpoints, re-signing only the MTProto obfuscation header. Full card: https://gitnova.dev/en/r/ToiCF/CF-Workers-TGProxy.md 4. **denoland/celld** — 4.1 · Breakout · DevOps & infrastructure · Rust · +11 stars measured 2026-10-11, 11:08–12:37 UTC, ≈ 44 by evening Open-source daemon that runs Cloudflare Workers applications on your own machines: Workers, Durable Objects, KV, Queues, D1, R2, Workflows, Cron and static assets from your existing wrangler.json. State lives in your own S3-compatible… Full card: https://gitnova.dev/en/r/denoland/celld.md 5. **cloudflare/workerd** — 4.0 · Breakout · DevOps & infrastructure · C++ · +10 stars measured 2026-10-11, 00:39–12:28 UTC, ≈ 21 by evening A JavaScript/Wasm server runtime based on Cloudflare Workers, letting you self-host, test, or proxy requests with Workers-style code. Aimed at developers deploying edge-style services outside the Cloudflare network. Full card: https://gitnova.dev/en/r/cloudflare/workerd.md --- Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score. Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast. Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself. Data as of 2026-10-11 12:38 UTC, updated every 30 minutes.