What tools does the llm-latency-tracker MCP server provide?

llm-latency-tracker exposes 2 tools: get_ai_api_latency, get_model_deprecations. This list was returned by the running server when we connected to it.

Does llm-latency-tracker require an API key?

No. llm-latency-tracker answered a tools/list request without credentials.

How do I install llm-latency-tracker?

Run: claude mcp add --transport http llm-latency-tracker https://llmlatency.dev/mcp. One-click install links for Cursor and VS Code are also available on the server's page.

Connection check verified live · 28h ago

llm-latency-tracker

Measured latency, time to first token and uptime for ~45 AI inference APIs, by region.

AI & LLM tooling Monitoring & observability Productivity & docs

Install

Remote (HTTP)

claude mcp add --transport http llm-latency-tracker https://llmlatency.dev/mcp

Tools 2

GitHub stars —

Installs / wk —

Licence —

Transport streamable-http

Last checked 28h ago

Tools & capabilities

2 tools

Read from the running server on 28h ago.

get_ai_api_latency region

Measured latency (TTFB p50/p95) and uptime rankings of AI inference API providers by region, from llmlatency.dev.

get_model_deprecations provider

AI model deprecation calendar: announced and shutdown dates, replacement models, and how many days of migration notice each provider actually gives (median/min/max). Every entry is…