LLM Leaderboard

Artificial Analysis snapshot plus carried benchmark scores for current frontier LLMs

Last updated: 2026-05-23 · current AA models (AA Index ≥ 25) · "me" = output t/s measured locally via OpenRouter

Reasoning · cost vs intelligence Pareto frontier · hover for details
Reasoning · throughput vs intelligence Pareto frontier · hover for details
Access
Provider
Type
Efficiency