What LLM Stats does

LLM Stats is an independent AI evaluation and benchmarking hub. It aggregates public benchmark results and live API metrics to rank and compare 300+ AI models across many dimensions, giving a vendor-neutral view of model performance, speed, price, and context length.

Key capabilities

  • Leaderboards across general performance, open-source models, coding, writing, math, reasoning, long context, tool calling, and image and video generation
  • Tracks intelligence via benchmarks such as GPQA and SWE-Bench, plus output throughput, pricing, and context window size
  • Side-by-side model comparisons
  • Continuously updated rankings, with pricing validated frequently and performance metrics on a rolling average

Who it's for

LLM Stats serves developers and engineers choosing models for specific tasks, AI product teams benchmarking against competitors, researchers needing standardized results, and cost-conscious teams looking for the best performance per dollar, all through transparent, continuously updated methodology rather than vendor marketing.