OmniMatBench: leaderboard
As multimodal language models play an increasingly important role in scientific research, materials science offers a critical testbed due to its interdisciplinary, multimodal, and application-driven nature. However, existing materials benchmarks mainly focus on property prediction, knowledge QA, or characterization understanding, leaving the broader reasoning process from materials knowledge to ap
Metric: Avg. Score (self-reported). Source: benchmarklist.com. Saturation forecast: Around June 2027. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.7 | 37.2 | #49 |
| 2 | Gemini 3.1 Pro (Preview) | 34.7 | #58 |
| 3 | GPT-5.5 | 32.4 | #31 |
| 4 | Qwen 3.6 Plus | 29.9 | #120 |
| 5 | Grok 4.3 | 29.1 | #188 |
| 6 | O3 | 28.8 | #124 |
| 7 | Gemma 4 31B | 25.7 | #200 |
| 8 | O4 Mini | 25.4 | #180 |
| 9 | Qwen 3 VL 235B A22B | 24.7 | #238 |
| 10 | GLM-4.6V | 21.2 | #316 |
| 11 | Llama 4 Maverick | 20 | #464 |
| 12 | GPT-4o | 14.6 | #343 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=omnimatbench · How It Works · Data refreshed daily, snapshot 2026-10-20.