CAE-Bench RC7 - Overall: leaderboard

Metric: Score (0-100), equal weight over 80 tasks. Source: arxiv.org. Saturation forecast: Around February 2027. 12 models tracked. Updated 18 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6.1 SolMedium65.24#12
2GPT-6 AstraMedium64.85#7
3Gemini 4 ArgonMedium62.49#14
4GPT-5.6 SolMedium60.23#33
5GPT-5.6 LunaMax59.32#129
6Gemini 3.8 FlashMedium58.3#39
7GPT-6 SolMedium57.93#30
8GPT-6 LunaxHigh56.35#102
9DeepSeek V4.1 FlashHigh50.41#67
10Qwen 3.8 FlashLow42.36#127
11Kimi K2.7 CodeThinking40.28#195
12GLM-5.3 FlashLow37.03#134

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=cae-bench-rc7-overall · How It Works · Data refreshed daily, snapshot 2026-11-18.