Arena Elo (Vision)

Arena Elo (Vision) — frequently asked questions

What is Arena Elo (Vision)?

Real people give two anonymous AIs the same image — a photo, a screenshot, a diagram — and vote for whichever reads it better. The votes become a chess-style Elo rating on arena.ai. It measures which AI people find more useful at looking at things, not how it scores on a fixed test set. Higher is better.

Which AI model scores highest on Arena Elo (Vision)?

Claude Fable 5 by Anthropic holds the best Arena Elo (Vision) result among tracked models, at 1312 (released Jun 9 2026). Higher scores are better on this benchmark.

What are the top 5 models on Arena Elo (Vision)?

1. Claude Fable 5 (Anthropic) — 1312; 2. Qwen3.8-Max (Qwen) — 1302; 3. Claude Opus 4.7 (Anthropic) — 1301; 4. Claude Opus 4.6 (Anthropic) — 1299; 5. Muse Spark (Meta) — 1294.

How many models have a published Arena Elo (Vision) score?

31 tracked models have a published Arena Elo (Vision) score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.

What is the best open model on Arena Elo (Vision)?

Qwen3.8-27B by Qwen is the highest-ranked model with downloadable weights on Arena Elo (Vision), scoring 1253 at rank 26 overall.