Skip to content
AIRank

About

About AIRank

AIRank is an independent, multilingual leaderboard of AI models. We rank models across chat, coding, image, video, agents, audio and search, combining public benchmarks, community signals and editorial judgment.

Methodology

Scores are computed from the community-voted LMArena Elo ratings, mapped to a 0–100 scale: 1000 Elo = 0, 1400 Elo = 100. When no arena rating is available, the Hugging Face trending score is used as a fallback. Rankings refresh daily. Sources: LMArena, Hugging Face, OpenRouter and Artificial Analysis.

Independence

AIRank is free and does not sell ranking positions. Sponsorships are clearly labelled and never influence scores.

Get in touch

Questions, corrections or partnership ideas? Write to us — we read everything.

Contact