OmniMatBench - QA - Materials Processing and Manufacturing: leaderboard
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 Materials Processing and Manufacturing 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 April 2028. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 38.4 |
| 2 | Gemini 3.1 Pro (Preview) | 34.6 |
| 3 | GPT-5.5 | 33.3 |
| 4 | O3 | 30.3 |
| 5 | Grok 4.3 | 29.4 |
| 6 | Qwen 3.6 Plus | 27.8 |
| 7 | Gemma 4 31B | 23.2 |
| 8 | O4 Mini | 21.8 |
| 9 | Qwen 3 VL 235B A22B | 20.9 |
| 10 | GLM-4.6V | 18.5 |
| 11 | GPT-4o | 15.2 |
| 12 | Llama 4 Maverick | 13.5 |
Interactive version: theaggregate.ai/benchmark?slug=omnimatbench-qa-materials-processing-and-manufacturing · How It Works · Data refreshed daily, snapshot 2026-10-07.