"wgpu" matching MCP servers:

Enables AI agents and their users to submit, route, and monitor training jobs across free-tier GPU clouds (Kaggle, Colab, Lightning, local MPS) through a local daemon that tracks quota, explains routing decisions, and gates long or secret-reading agent jobs behind human approval.

An MCP stdio server that lets an AI agent run ComfyUI or VoiceStudio batch jobs on a GPU shared with a local LLM server by unloading the LLM, launching the executor, running the jobs, and always reloading the LLM afterward. It exposes comfy_batch, voice_batch, and gate_status tools so a single blocking call handles the whole GPU hand-off safely even if the client disconnects.

Exposes NVIDIA GPU metrics (info, utilization, VRAM, temperature) via MCP tools for real-time querying from AI assistants.

Enables recording a Wine or macOS program on an Apple silicon Mac and analyzing its GPU usage — GPU milliseconds per frame, frame pacing, FPS and memory — through Instruments' Metal System Trace, and also reads Metal .gputrace captures without Xcode and reports Apple's published GPU family facts. Recording, listing, comparing, and paging through traces and captures are exposed as tools that a shell or an AI coding agent can call directly.

Exposes local NVIDIA GPU (CUDA) and Rust-to-WASM toolchain as MCP tools for sovereign local compute.

Let me evaluate the README content.

MCP server for AI-controlled GPU VRAM management in ComfyUI, allowing Claude to monitor and control GPU state, manage model offloading, and start/stop ComfyUI.

Enables AI agents to submit GPU job requests, select eligible workers, track execution, and retrieve results via MCP.

Provides a persistent compute queue for AI agents, enabling them to submit hardware-aware GPU jobs, monitor execution, and retrieve results and artifacts across sessions via MCP tools.

Enables interaction with Hyperbolic's GPU cloud, allowing users to view available GPUs, rent instances, establish SSH connections, and run GPU-powered workloads through natural language commands.

Enables MCP clients to discover NVIDIA, AMD, and Intel GPUs, monitor utilization, memory, temperature, power, clocks, and processes, retain bounded local history, and estimate whether AI models fit available VRAM before loading them.

Enables approved submission, monitoring, and retrieval of Kaggle CPU/GPU jobs, including two-stage preparation-to-training workflows, experiment comparison, artifact/log access, and SMTP completion notices.

An MCP server that enables AI agents to query real-time NVIDIA GPU metrics like utilization, memory, temperature, and power without external monitoring tools.

Enables AI coding agents to record auditable work ledgers with evidence chains, from contract to proof packet, via MCP tools for file scanning, code review, and issue triage.

A free, open-source point-and-click web scraper that connects your browser to Claude over the Model Context Protocol, allowing you to highlight elements on any page and scrape them without manual copying.

Provides read-only visibility into local CPU pools and Apple GPU lane usage inside Claude Code and Codex.

Enables AI coding agents to run policy-constrained Python jobs on lab NVIDIA GPU hosts, including GPU inspection, device reservation, job launching and monitoring, and owner-scoped process stopping.

Provides 19 tools for discovery, data analysis, configuration, and simulation, enabling an agent to drive digital twin workflows with session persistence and knowledge graph capabilities.