Spatial Competence Benchmark - Constructive: leaderboard

The 171 constructive subtasks of the Spatial Competence Benchmark (SCBench, 2026), such as constructive solid geometry, projections, packing and triangulation, with executable outputs checked by deterministic verifiers, single-turn and without tools. Accuracy in percent; higher is better.

Metric: Accuracy (%) over the 171 constructive subtasks (CSG, projections, packing, triangulation, curve and loop fitting and similar construction tasks) of the Spatial Competence Benchmark (SCBench) subtasks with executable outputs checked by deterministic verifiers or simulators, each subtask scored in [0, 1] (binary or partial credit), every subtask weighted equally; single-turn, zero-shot, one sample, no tools, no self-correction; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 3 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1GPT-5.2xHigh51.9#190
2Gemini 3 Pro (Preview)51.4#123
3Claude Sonnet 4.5Thinking30.2#267

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

Interactive version: theaggregate.ai/benchmark?slug=spatial-competence-benchmark-constructive · How It Works · Data refreshed daily, snapshot 2026-11-22.