OmniMatBench - Calculation: 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 strict exact-slot accuracy of the calculation answers, macro-averaged over the subfields; percent, higher is better.
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 October 2028. 13 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 29.1 | #85 |
| 2 | Gemini 3.1 Pro (Preview) | 28.7 | #99 |
| 3 | GPT-5.5 | 25.5 | #54 |
| 4 | Qwen 3.6 Plus | 24.9 | #238 |
| 5 | Intern-S1-Pro | 23.2 | #585 |
| 6 | Grok 4.3 | 22.9 | #342 |
| 7 | Qwen 3 VL 235B A22B | 20.7 | #485 |
| 8 | O4 Mini | 20.1 | #368 |
| 9 | O3 | 19.9 | #237 |
| 10 | Llama 4 Maverick | 19.3 | #1110 |
| 11 | Gemma 4 31B | 19.2 | #382 |
| 12 | GLM-4.6V | 17.4 | #710 |
| 13 | GPT-4o | 7.2 | #751 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=omnimatbench-calculation · How It Works · Data refreshed daily, snapshot 2026-11-21.