Epoch AI - Ebr Bench: leaderboard

Metric: Score. Source: epoch.ai. 21 models tracked.

Top models

#ModelScore
1GPT-6 (Max)76.19
2Claude Opus 5 (Max)50
3Claude Fable 5.1 (Max)47.62
4Claude Fable 5 (Max)39.52
5GPT-5.6 Sol (Max)39.05
6GPT-5.5 (xHigh)34.29
7Grok 4.6 (xHigh)30.48
8Claude Opus 4.8 (Max)28.57
9GPT-5.4 (xHigh)25.4
10GPT-5.2 (xHigh)23.02
11Claude Opus 4.7 (Max)19.05
12Gemini 3.1 Pro (Preview)14.29
13GPT-5 (High)12.7
14Claude Opus 4.6 (Max)12.7
15Qwen 3.7 Max9.52

Interactive version: theaggregate.ai/benchmark?slug=epoch-ai-ebr-bench · How It Works · Data refreshed daily, snapshot 2026-09-05.