mcpcost

io.github.pmady/gpu-mcp-server

GPU MCP Server

Is the GPU MCP server down? Yes. It has given no valid answer since 2026-10-29: crashed on start.

NVIDIA GPU metrics as MCP tools — utilization, memory, temperature, power. Supports MIG.

crashed on start down since 2026-10-29 · probed 2026-11-14 · unchanged since 2026-11-12

MCP error -32000: Connection closed
time=2026-10-29T07:49:14.224Z level=INFO msg="nvml: nvml init: ERROR_LIBRARY_NOT_FOUND"
Run it
docker run -i --rm ghcr.io/pmady/gpu-mcp-server:v0.1.0 (stdio, probed in a container with the network off)
Host
ghcr.io
Repository
https://github.com/pmady/gpu-mcp-server
Registry
active, updated 2026-06-25 · registry entry

Compare this server with others.

More servers

Other registry entries on ghcr.io, the ones next to this name; all of them.

Servers that mention temperature or utilization.

Badge for the README

mcpcost: crashed on start

The result of the latest probe, the tool count and the token cost. The feed has an entry for every probe that changed them.

Atom feed

A mail when this server changes

Stops answering, comes back, tool set or cost changes. One confirmation mail first, a stop link in every mail.

For the publisher