inferbench for Claude Code

io.github.rudrendupaul/inferbench

Benchmarks local LLM inference speed (tokens/sec) on your own hardware via MCP tools.

client:Claude Code transport:stdio runtime:pypi

Install inferbench in Claude Code

run in your project directory

claude mcp add inferbench -- uvx inferbench-cli

Adds it for this project only. Append --scope user to make it available everywhere. Claude Code docs