Runs a throughput benchmark for LLM APIs, measuring generation throughput, prompt throughput, and Time To First Token (TTFT) under various concurrency levels.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"llm-api-benchmark-mcp-server","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# LLM API Benchmark MCP Server This project provides an MCP (Model Context Protocol) server designed to benchmark Large Language Model (LLM) APIs. It allows you to measure various performance metrics such as generation throughput, prompt throughput, and Time To First Token (TTFT) for your LLM endpoints. ## Features * **Comprehensive Benchmarking:** Measure key performance indicators like generation throughput, prompt throughput, and Time To First Token (TTFT). * **Flexible Deployment:** Supports both remote SSE server for quick trials and local deployment via stdio or SSE transport. * **Customizable Benchmarks:** Configure various parameters for your LLM API benchmarks, including concurrency levels, model names, token limits, and more. * **Detailed Output:** Provides structured JSON output with aggregated and distributed metrics for in-depth analysis. * **Easy Integration:** Designed as an MCP server for seamless integration with MCP clients. ## Table of Contents - [LLM API Benchmark MCP Server](#llm-api-benchmark-mcp-server) - [Features](#features) - [Table of Contents](#table-of-contents) - [Prerequisites](#prerequisites) - [Install uv](#install-uv) - [Quick Start](#quick-start) - [Demo (Recommended for Simple Trials)](#demo-recommended-for-simple-trials) - [Local Deployment (stdio)](#local-deployment-stdio) - [Local Deployment (sse)](#local-deployment-sse) - [Usage Example](#usage-example) - [Benchmark Parameters](#benchmark-parameters) - [License](#license) ## Prerequisites This project uses [`uv`](https://docs.astral.sh/uv/getting-started/installation/#installation-methods) for environment management. Please ensure you have `uv` installed on your system. ### Install uv Refer to the [official `uv` documentation](https://docs.astral.sh/uv/getting-started/installation/#installation-methods) for installation methods. ## Quick Start You can experience this MCP server through a remote SSE server for quick trials or …
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.