A GPU task manager for Windows, Linux, and WSL: per-process VRAM, GPU activity, and LLM model/KV-cache insights (Windows AMD and NVIDIA).
Python
Commits on main
feat: adding support for AMD, tested with Vulkan and ROCm/HIP processes (ex. llama) on a Radeon 9060XT 16GB, Windows.
dbf61a8 K committed 16 days ago
Tighten WSL attribution evidence and first-sample reporting
d56e44a Scott Hanselman committed 16 days ago
Keep WSL VRAM matches honest and learn idle contexts at startup
3338e0b Scott Hanselman committed 16 days ago
Keep the Linux launcher executable when the tarball is built on Windows
02f6039 Sameh Khamis committed 17 days ago
Add eight-chart responsive TUI, video clock, and portable x64/ARM64 release builds
f37e8ad Scott Hanselman committed 17 days ago
Show Handy's selected model and runtime-reported model events with explicit provenance
0c1f757 Scott Hanselman committed 17 days ago
Add cautious local decision-model hints and document GPU-only scope
b2b3186 Scott Hanselman committed 17 days ago
Identify loaded models by evidence, cheaply, across runtimes
f2a7492 Scott Hanselman committed 17 days ago
Add GPU introspection TUI with documented quick start and screenshots
d793e24 Scott Hanselman committed 17 days ago