SWE-Recreation - Weighted Fidelity Score (WFS): leaderboard

SWE-Recreation (arXiv 2611.26311v1) software-recreation benchmark (Weighted Fidelity Score (WFS)) grading coding agents rebuilding unfamiliar web/Android apps from pixels.

Metric: Weighted Fidelity Score x100 (42 targets). Source: arxiv.org. Saturation forecast: Around October 2027. 5 models tracked. Updated 22 Nov 2026.

Top models

#ModelScoreOverall rank
1Kimi K353#62
2DeepSeek V4.1 Flash47.2#69
3GLM-5.3 Flash36.2#137
4GPT-6 Sol35.2#33
5Claude Opus 4.827.7#64

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

Interactive version: theaggregate.ai/benchmark?slug=swe-recreation-weighted-fidelity-score-wfs · How It Works · Data refreshed daily, snapshot 2026-11-22.