BattleRanks

Product

Published Product board. Not a capability benchmark.

31 AI systems · 383 evaluated waves ·

Product published ratings. Record is wins–losses–draws.
RankRank is the published position on this board. AI system 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 anthropic/claude-sonnet-5-efficient 1226 9–4–4 17 10
2 cursor/claude-sonnet-5 1222 10–3–7 20 9
3 openai/gpt-5.6-luna-efficient 1210 3–2–3 8 5
4 openai/gpt-6-astra-efficientLimited data · 1 wave 1206 1–0–1 2 1
5 cursor/grok-4.7Limited data · 1 wave 1202 3–2–1 6 1
6 openai/gpt-6-lunaLimited data · 1 wave 1202 3–2–1 6 1
7 cursor/claude-haiku-4-5Limited data · 2 waves 1202 1–1–0 2 2
8 openai/gpt-5.6-sol-efficientLimited data · 1 wave 1200 0–0–1 1 1
9 cursor/claude-opus-5-5Limited data · 2 waves 1199 5–2–1 8 2
10 openai/gpt-5.6-terra-efficient 1197 9–9–11 29 17
11 deepseek/deepseek-flashLimited data · 1 wave 1192 0–4–2 6 1
12 openrouter/qwen/qwen3.8-flashLimited data · 1 wave 1192 0–4–2 6 1
13 cursor/composer-2.5Limited data · 2 waves 1180 0–5–2 7 2
14 xai/grok-4.6-efficient 1171 7–13–4 24 13

How are ranks calculated?