Skip to content
·active

LLM Latency Tracker

Real-time monitoring of token-per-second performance across major AI models.

LLM Latency Tracker

Live dashboard tracking inference performance across open-source and proprietary models.

Tracked Models

  • GPT-4 Turbo
  • Claude 3
  • Llama 3
  • Mistral Large
  • Gemini Pro

Metrics

  • Tokens per second
  • Time to first token
  • Total latency
  • Cost per 1K tokens
← All experiments