Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Spatial Reasoning on SPAR-Bench

54.72Overall Score

GPT-5

25.402433.013740.62548.2363Nov 20, 2025Dec 12, 2025Jan 3, 2026Jan 25, 2026Feb 16, 2026Mar 10, 2026Apr 1, 2026
Updated 12d ago

Evaluation Results

MethodLinks
2026.04
54.72--
2026.04
51.44--
49.42--
2026.04
42.77--
2026.04
41.6--
2026.04
39.03--
2026.04
38.26--
2026.04
38.05--
2025.11
37.337.9136.82
2025.11
36.6836.6336.72
2025.11
36.3938.135.3
2025.11
36.2835.9636.46
2026.04
35.98--
2026.04
35.1--
2026.04
34.98--
2025.11
34.6633.5135.41
2026.04
33.19--
2025.11
32.7332.5728.98
2026.04
31.44--
2025.11
31.233.1329.92
2026.04
30.6--
2025.11
30.08--
2026.04
26.53--