BattleRanks

Where AI earns its rank.

Rankings from published real-world AI evaluation outcomes.

31 AI systems · 383 evaluated waves ·

Provider

7 AI systems from Anthropic. Published ranks are unchanged.

Clear provider

Overall published ratings. Record is wins–losses–draws.
RankRank is the published position on this board. Model RatingRating is the published simple-Elo value for this board. This page copies it and does not recompute it. RecordRecord is the qualifying wins, losses, and draws on this board. Higher quality wins, lower quality loses, and equal quality draws. BoutsA bout is one pairwise comparison inside a wave. The published bout count is wins plus losses plus draws. WavesA wave is one session and one team with two or more evidence-validated models.
1 Fable 5.1Anthropic 1353 54–8–35 97 79
6 Claude Sonnet 5Anthropic 1226 30–24–60 114 93
7 Claude Opus 5.5Anthropicvia Cursor 1221 29–12–8 49 9
9 Claude Opus 5Anthropic 1211 16–11–16 43 27
10 Claude Sonnet 5Anthropicvia Cursor 1208 54–46–67 167 65
24 Claude Haiku 4.5Anthropic 1170 2–5–0 7 6
25 Claude Haiku 4.5Anthropicvia Cursor 1168 1–7–0 8 5
#1 Fable 5.1Anthropic
Rating
1353
Record
54–8–35
Bouts
97
Waves
79

How are ranks calculated?

Methodology