OmniMatBench - Calculation: leaderboard
Metric: Calculation strict exact-slot accuracy (%): share of structured answer slots (values, units, formulas) exactly matching the reference after normalization, macro-averaged over the 19 subfields, on the 1,910 calculation questions 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 2029. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.7 | 29.1 |
| 2 | Gemini 3.1 Pro (Preview) | 28.7 |
| 3 | GPT-5.5 | 25.5 |
| 4 | Qwen 3.6 Plus | 24.9 |
| 5 | Grok 4.3 | 22.9 |
| 6 | Qwen 3 VL 235B A22B | 20.7 |
| 7 | O4 Mini | 20.1 |
| 8 | O3 | 19.9 |
| 9 | Llama 4 Maverick | 19.3 |
| 10 | Gemma 4 31B | 19.2 |
| 11 | GLM-4.6V | 17.4 |
| 12 | GPT-4o | 7.2 |
Interactive version: theaggregate.ai/benchmark?slug=omnimatbench-calculation · How It Works · Data refreshed daily, snapshot 2026-10-07.