SPUR - Morphological Perception: leaderboard

SPUR (2026) tests multimodal models with multiple-choice questions on multi-panel biomedical experimental figures from PubMed Central. Accuracy in percent on morphological perception questions, which ask the model to analyze cell shape and tissue structure in stained preparations; higher is better.

Metric: Accuracy (%) on SPUR the 634 morphological perception multiple-choice questions about multi-panel biomedical experimental figures from PubMed Central papers (questions drafted by GPT-4o, kept only when GPT-4o failed them in at least six of ten text-only attempts, then expert-reviewed): the questions ask the model to analyze cell shape, tissue architecture and other structures in stained preparations; higher is better. Source: arxiv.org. Saturation forecast: Around June 2027. 19 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3 Pro (Preview)67.74#119
2O4 MiniHigh64.5#368
3Claude 3.7 SonnetThinking64.32#470
4Seed-1.663.51#469
5Gemini 2.5 Pro (Preview 06-05)62.97#213
6GLM-4.5V61.99#754
7GPT-5.161.72#240
8Ministral-3-14B-Instruct-251261.4#1639
9Llama 4 Maverick59.78#1110
10Qwen 3 VL 30B A3B (Thinking)58#719
11Ministral-3-8B-Instruct-251257.03#1907
12Grok 4.1 Fast55.99#427
13Qwen 3 VL 30B A3B Instruct53.31#895
14InternVL3-78B51.97#759
15Mistral Small 3.151.92#1706

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=spur-morphological-perception · How It Works · Data refreshed daily, snapshot 2026-11-21.