CLASHarness v8 - Pass Rate (Mid): leaderboard

Metric: Episode pass rate, mid episodes (%). Source: github.com. Saturation forecast: Around 2030. 12 models tracked. Updated 18 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraHigh28.1#7
2Claude Opus 5High25#24
3GPT-5.6 SolMedium24.6#33
4Claude Opus 4.8Medium24.6#62
5Gemini 3.7 FlashMedium24.1#38
6Kimi K3High23.2#60
7GPT-5.6 LunaHigh22.3#129
8Gemini 3.6 FlashMedium21.9#77
9Claude Sonnet 5High21.4#116
10GPT-5.6 TerraHigh20.1#69
11Gemini 3.5 FlashLow14.3#106
12Gemini 3.1 Pro (Preview)High12.1#99

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

Interactive version: theaggregate.ai/benchmark?slug=clasharness-v8-pass-rate-mid · How It Works · Data refreshed daily, snapshot 2026-11-18.