Developer builds independent AI API latency tracker covering 45 providers across 4 regions

A developer created LLM Latency Tracker, an open-source tool that measures AI inference API latency from four global regions. It covers 45 providers including Western and Chinese labs, tracking both edge latency and time-to-first-token. The tool is designed to be agent-readable via MCP server and structured data formats.