CLASHarness v8 - Pass Rate (Hard): leaderboard

Metric: Episode pass rate, hard episodes (%). Source: github.com. Saturation forecast: Around 2029. 12 models tracked. Updated 18 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraHigh18.6#7
2GPT-5.6 SolHigh13.6#33
3Claude Opus 5High13.6#24
4Claude Opus 4.8Medium12.9#62
5Kimi K3High12.1#60
6GPT-5.6 LunaLow12.1#129
7GPT-5.6 TerraMedium10#69
8Gemini 3.7 FlashLow9.3#38
9Gemini 3.6 FlashLow9.3#77
10Claude Sonnet 5High7.9#116
11Gemini 3.1 Pro (Preview)Medium6.4#99
12Gemini 3.5 FlashLow5.7#106

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-hard · How It Works · Data refreshed daily, snapshot 2026-11-18.