OmniMatBench - QA - Fundamental Materials Science: leaderboard

OmniMatBench (2026) poses 3,171 expert-curated multimodal materials-science problems, open-ended questions graded against expert key points and calculations graded slot by slot, across 19 subfields. This board is the QA macro-F1 on the Fundamental Materials Science group, 0 to 100; higher is better.

Metric: QA macro-F1 (0-100): F1 between the response and the expert key points (precision and recall judged by Gemini 2.5 Flash), macro-averaged over subfields, times 100, over the QA questions of the Fundamental Materials Science group on OmniMatBench (3,171 expert-curated multimodal materials-science problems in 19 subfields: 1,261 open-ended QA and 1,910 calculation questions); higher is better. Source: arxiv.org. Saturation forecast: Around October 2027. 13 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Claude Opus 4.754#85
2Gemini 3.1 Pro (Preview)52.1#99
3O350.3#237
4Grok 4.348.9#342
5GPT-5.548.2#54
6O4 Mini45.8#368
7Qwen 3.6 Plus45.5#238
8Qwen 3 VL 235B A22B42.6#485
9Gemma 4 31B42#382
10Intern-S1-Pro38.8#585
11GLM-4.6V36.5#710
12GPT-4o33.9#751
13Llama 4 Maverick33.8#1110

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

Interactive version: theaggregate.ai/benchmark?slug=omnimatbench-qa-fundamental-materials-science · How It Works · Data refreshed daily, snapshot 2026-11-21.