Gguf MCP runs on your own machine — the client starts it, so there is no endpoint to ping. 56 installs a week from npm. Last commit 15 Aug 2026.
Inspect local GGUF/safetensors models: quantization, params, VRAM fit — headers only.
Today is the operative word: we check Gguf MCP every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add gguf-mcp -- npx -y gguf-mcp
{
"mcpServers": {
"gguf-mcp": {
"args": [
"-y",
"gguf-mcp"
],
"command": "npx"
}
}
}
[mcp_servers.gguf-mcp]
command = "npx"
args = ["-y", "gguf-mcp"]
{
"mcpServers": {
"gguf-mcp": {
"args": [
"-y",
"gguf-mcp"
],
"command": "npx"
}
}
}
{
"mcpServers": {
"gguf-mcp": {
"args": [
"-y",
"gguf-mcp"
],
"command": "npx"
}
}
}
Read-only ops/health MCP server for a local vLLM instance: health tiers, GPU/VRAM, service status.
Multi-model AI gateway — 8 providers plus local models, no system prompt injection.
MCP server for SQL Server: schema inspection, safe read-only queries, query plan analysis
Local quantum error correction tools; decoder output verified against H c = s (mod 2).
Inspect and manage local files and folders inside allowed roots on macOS.
Local MCP gateway daemon: route, audit, and inspect MCP servers from one config.
Persistent file-based memory for AI agents: inspectable, versionable, fully local.
Benchmark local LLM models — speed, quality & hardware fitness verdict from any MCP client
Answers built from our own checks of this server.