List your company

Spatial Reasoning on ReSQ

85.41GPT-grid ScoreGPT-5.1
68.187672.658877.1381.6012Jun 30, 2026

Evaluation results

MethodLinks
GPT-5.1Setting=FG (Full-grid)2026.0685.41—
Qwen3-32BSetting=PG (Pruned-grid)2026.0685.2579.83
GPT-5.1Setting=PG (Pruned-grid)2026.0684.59—
Qwen3-32BSetting=FG (Full-grid)2026.0683.2879.5
GPT-5.1Setting=Text2026.0682.6—
LLaMA3.1-70BSetting=FG (Full-grid)2026.0680.3378.2
Qwen3-32BSetting=Text2026.0680.380.3
LLaMA3.1-70BSetting=PG (Pruned-grid)2026.068078.2
GPT-5.1Setting=Rel. Trip.2026.0679.18—
LLaMA3.1-70BSetting=Text2026.0677.0577.05
LLaMA3.1-70BSetting=Rel. Trip.2026.0676.2369.84
Qwen3-32BSetting=Rel. Trip.2026.0668.8568.85