benchgap
Calibration

VulcanBench v3 → CursorBench 3.2

CursorBench 3.2 is estimated from VulcanBench v3 with a Michaelis–Menten curve fitted on 8 models measured on both: y = 1.8682·x / (1.55108 + x), R² = 0.43, cross-validated error 3.3 pp. It is used for 6 estimates.

Estimated modelVulcanBench v3CursorBench 3.2Source
Claude Haiku 4.576.2%61.5%estimated ± 3.3 pp, medium confidence
DeepSeek V4 Flash 073188.4%67.8%estimated ± 3.3 pp, medium confidence
GLM-5.378.3%62.7%estimated ± 3.3 pp, medium confidence
Muse Spark 1.287.0%67.1%estimated ± 3.3 pp, medium confidence
Qwen3.8-27B82.6%64.9%estimated ± 3.3 pp, medium confidence
Qwen3.8 Max81.2%64.2%estimated ± 3.3 pp, medium confidence