Tools & ProjectsNovember 12, 2026
50
š¶ļøš¶ļøš¶ļø
Microsoft VP ships GPU task manager
GPU Taskman displays per-process VRAM usage and detects local LLMs.
#GPU#open-source#LLM#monitoring#VRAM

š„ What happened
Scott Hanselman, Microsoft VP, released GPU Taskman ā an open-source task manager for GPUs. It runs on Windows, Linux (including WSL), and macOS on Apple Silicon.
š” Why it matters
Beyond utilization and temperature, it shows which AI models a process uses ā including parameter count and KV-cache estimates. For anyone running local LLMs, it's the overdue X-ray view into VRAM.
ā” Our take
Finally you can see which model is eating your VRAM. But beware: model attribution is heuristic, not measurement ā read the release notes before trusting it.
The title, summary and analysis of this item were produced automatically by an AI system and have not been editorially reviewed. They may contain errors, bias or omissions ā when in doubt, read the linked original source.
Deep Dives & Similar Intelligence
Context Windows Obsolete: Billion-Token Sessions
Tools & ProjectsNovember 11, 2026
OpenMausBot: Your Own Local Agent Team
Tools & ProjectsNovember 08, 2026

MCP Burns 59K Tokens Per Task
Tools & ProjectsNovember 11, 2026

145B Model Runs on a Laptop
AI modelsNovember 12, 2026

Google's Carbon: Frontier Model for Agent Swarms
AI modelsNovember 10, 2026