MCP server

llm-d-benchmarking-agent

Chat-based assistant that benchmarks llm-d from a plain-English goal: it interviews you, deploys an llm-d stack if needed, drives the real llm-d-benchmark CLI in a security sandbox, and explains the results. Also a Kubernetes run orchestrator, results analyzer, capacity pre-flight, shareable reports, and an MCP server.

• Published source
• Repository

Version

main

Last observed

11 October 2026

This is public metadata, not a safety certification or complete assessment. Static analysis cannot establish what a tool does at runtime.