Benchmark RadarAI BENCHMARK PROFILE
OmniMatBench
OmniMatBench assesses multimodal reasoning in materials science across 19 subfields with 3,171 expert-curated QA and calculation problems, spanning four domains from fundamental knowledge to applied materials.
- Released
- 2026-05-28
- Readiness
- Paper only
- Primary field
- Science & Research
Why it matters
Existing materials benchmarks focus on narrow tasks; OmniMatBench provides a broad reasoning benchmark revealing a substantial gap in current MLLMs, guiding AI assistant development in materials research.
Motivation
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.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.