gputaskman: an open-source htop for your GPU that breaks down VRAM by process and LLM KV-cache
alex_verem · x · 2026-11-17
Task Manager only tells you "15GB used" — gputaskman tells you by what.
- Shows per-process GPU memory share and detects loaded AI models
- Splits an LLM's footprint into model weights vs KV cache (conversation memory), so you can see why a model won't fit or which browser tab is stealing VRAM
- Works on Windows, Mac and Linux, free and open source
The author calls it the first thing to check before buying a bigger graphics card.
Related event: Open-Source gputaskman Gains Traction as htop for GPUs(2 posts)→
More from coding & agent
- Gemini 4 Carbon vibe check: boldest UI designs, least reliable finishes — every · 2026-11-17
- Matt Pocock's /constrain Idea: Cage Agents With Brutal Lint Rules — mattpocockuk · 2026-11-17
- Editing WordPress via MCP Feels Like "Superpowers"; Contributors Want It in Core — VeryWellVersed · 2026-11-17
- Self-hosted MCP endpoint exposes 38 database engines, read-only, creds stay server-side — Green_Competition_21 · 2026-11-17
- Stop Over-Trusting AI Coding Leaderboards: CACM Paper Cites OpenAI's SWE-bench Verified Reversal — EhudReiter · 2026-11-17
- SubCast: open-source bilingual podcast transcripts with per-sentence AI Q&A — vista8 · 2026-11-17