benchgap
Calibration

Vals SWE-bench → CursorBench 4.0

CursorBench 4.0 is estimated from Vals SWE-bench with a inverse Michaelis–Menten curve fitted on 8 models measured on both: y = 0.46732·(x − 0.0000) / (0.0000 + 2.0000 − x), R² = 0.46, cross-validated error 4.4 pp. It is used for 6 estimates.

Estimated modelVals SWE-benchCursorBench 4.0Source
Claude Opus 4.782.0%32.5%estimated ± 4.4 pp, medium confidence
Claude Sonnet 4.677.4%29.5%estimated ± 4.4 pp, low confidence
Gemini 3.1 Flash-Lite62.8%21.4%estimated ± 4.4 pp, low confidence
Gemini 3 Flash75.0%28.0%estimated ± 4.4 pp, low confidence
GPT-5.2-Codex72.4%26.5%estimated ± 4.4 pp, low confidence
Qwen3.5 Flash64.4%22.2%estimated ± 4.4 pp, low confidence