⬢github Python · 2.2K ★ +1.2K since we first saw it · pushed 1 h ago · MIT
shanselman/gputaskman
A GPU task manager for Windows, Linux, and WSL: per-process VRAM, GPU activity, and LLM model/KV-cache insights (Windows AMD and NVIDIA).
gputaskman is a terminal UI, like htop but for GPU internals, showing per-process VRAM usage, GPU engine activity, and—when it can—LLM model and KV-cache memory breakdowns on Windows (NVIDIA/AMD), Linux/WSL, and Apple Silicon Macs. It answers not just how much GPU memory is used, but by what: model weights, caches, or the browser.
Why now: Trending as a newly released, fast-starred project (nearly 1,000 stars within days) from a well-known developer, riding interest in local LLM inference and GPU memory visibility.
Who it is for: Developers and enthusiasts running local LLMs, training, or GPU-heavy workloads who want to see exactly which processes and models are consuming VRAM.
Stars over our 323 snapshots: 969 to 2.2K, since 3 d ago.
Where people talked about it
- ⬢github new repos, most starred 8 min ago
- ⬢github new repos, most starred 7 h ago
- ⬢github new repos, most starred 1 d ago
- ⬢github new repos, most starred 2 d ago
- ⬢github new repos, most starred 3 d ago
API: https://socialmediatrends-api.osmike.com/v1/repos/shanselman/gputaskman