Evolving Requirements - Evo StrictPass: leaderboard

61 frozen repository tasks, 207 requirement states. ClaudeCode2.1.74, 200ktoken context, high reasoning, provider-default sampling, network-disabled. Full one complete session; Evo one response/state; Evo+Comp initial plus<=2completion revisions/state; Hy4-Preview high feedbackgenerator20ktokens and<=2publicobservations/message. StrictPass allchecks; NodeCredit partialstate credit; FDAC feedback-discounted adaptivecorrectness. Pooled row withheld. Metric: Native score (%). Source: https://arxiv.org/html/2610.35488v1

Metric: Native score (%). Source: arxiv.org. Saturation forecast: Around December 2027. 12 models tracked. Updated 17 Nov 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-6 AstraHigh68.9#8
2GPT-5.6 SolHigh62.3#33
3DeepSeek V4 FlashHigh57.4#265
4DeepSeek V4.1 FlashHigh57.4#69
5Kimi K3High54.1#59
6Hy4 previewHigh52.5#75
7GLM-5.3High50.8#73
8DeepSeek V4 ProHigh49.2#198
9GLM-5.3 FlashHigh49.2#136
10Qwen 3.8 FlashHigh47.5#128
11Grok 4.6High37.7#47
12Hy3High32.8#211

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

Interactive version: theaggregate.ai/benchmark?slug=evolving-requirements-evo-strictpass · How It Works · Data refreshed daily, snapshot 2026-11-17.