Agent♥︎Age
Catalog

LLM Latency Tracker

Official1 toolLive

by mazamaka · Python

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

dev.llmlatency/llm-latency-tracker MCP Server

The dev.llmlatency/llm-latency-tracker MCP server provides access to measured latency, including time to first token and uptime for approximately 45 AI inference APIs by region. The project is described as independent and provider-neutral, and emphasizes that metrics are measured rather than scraped.

🛠️ Key Features

  • Latency measurement by region for ~45 AI inference APIs
  • Tracks time to first token and uptime
  • Associated tooling count: 1

🚀 Use Cases

  • Compare AI inference performance across regions and providers
  • Monitor operational status via uptime alongside latency metrics

⚡ Developer Benefits

  • MCP integration listed alongside a JSON API endpoint
  • Developer-focused stack topics include python, observability, and benchmark
  • Related resources: llmlatency.dev rankings, rankings.json, and a deprecations calendar

⚠️ Limitations

  • Coverage is limited to the ~45 inference APIs referenced in the description
  • No additional MCP tool details are provided beyond toolCount: 1

Topics

aianthropicbenchmarklatencyllmllms-txtmcpobservabilityopenaipython