Microsoft's Scott Hanselman releases open-source GPU Taskman tool on GitHub
Microsoft VP Scott Hanselman has published gputaskman, an open-source task manager that shows per-process GPU usage and VRAM consumption on Windows, Linux, and macOS. The tool, now at version 0.3.0, also breaks out AI and LLM-specific details such as model weights, KV cache size, and parameter counts for running processes.
GoKawiil's interpretation of the reporting above, not reported fact.
The tool addresses a gap in standard operating system utilities, which rarely expose granular GPU memory allocation the way Task Manager does for CPU and RAM. Its focus on surfacing AI/LLM-specific metrics suggests growing developer demand for visibility into how local AI workloads consume limited VRAM, particularly as more users run models like Qwen locally.
- Scott Hanselman released gputaskman, an open-source GPU task manager on GitHub.
- The tool works across Windows, Linux, and macOS and shows per-process VRAM and GPU usage.
- It includes detailed breakdowns for AI/LLM tasks, such as model weights and KV cache size.
NVIDIA GeForce RTX 4080 Super GPU — This is the exact GPU Hanselman used while testing GPU Taskman, making it a great reference point if you're curious about monitoring your own VRAM usage during AI and LLM workloads. With ample memory and strong performance, it's well suited to the kind of heavy multitasking this new tool is designed to track.
See NVIDIA GeForce RTX 4080 Super GPU on Amazon → Affiliate link — we may earn a commission on purchases, at no extra cost to you. Product picked by AI based on this article; it is not a tested recommendation.Source: tomshardware.com — Mark Tyson, 2026-11-08
Published there as: “Open-source GPU task manager reveals exactly what's eating your VRAM”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.