▲ 245 today1.5k starsPythonNew · 7d old
GPU task manager shows per-process VRAM use, not just total
What is gputaskman?
gputaskman is a terminal UI that lists how much GPU memory each process is using and what engine activity it has. It also tries to match that usage to LLM model files, weight tensors, and KV-cache buffers.
Who it is for
If you run AI models locally, this lets you pinpoint which process or model is hogging VRAM so you can optimise or debug.
Try it in 60 seconds
You need: Python 3.11+, Windows/Linux/WSL/macOS with a supported NVIDIA, AMD, or Intel GPU and drivers
uv tool install gputaskmangputaskman
MIT licenseUpdated todayPython8 open issues
New projects are not vetted. Skim the code, and run it in a fresh folder or a container before giving it your files or keys.
The story so far
Early Signal first spotted shanselman/gputaskman on November 9, 2026, when it had 1k GitHub stars. It has since grown to 1.5k. On its busiest day it gained 245 stars. The project is 7 days old and written mainly in Python. It is published under the MIT license and has 73 forks. Early Signal files it under Inference, LLMs, RAG & Memory.
Questions about gputaskman
- Is gputaskman free and open source?
- Yes. shanselman/gputaskman is open source under the MIT license, and the code is on GitHub at https://github.com/shanselman/gputaskman.
- How do I install gputaskman?
- Use one of these, depending on your setup: uv tool install gputaskman · gputaskman
- What do I need to run gputaskman?
- Python 3.11+, Windows/Linux/WSL/macOS with a supported NVIDIA, AMD, or Intel GPU and drivers