math benchmark
EMMA
EMMA (Enhanced MultiModal reAsoning) is a benchmark for organic multimodal reasoning across mathematics, physics, chemistry, and coding.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
EMMA leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 79.3% | 100.0% | 2 | C | |
| 2 | BY | 78.4% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
EMMA highlights
The top published results on this benchmark's own scale.
What is EMMA?
Definition and scoring fields from the benchmark registry.
EMMA (Enhanced MultiModal reAsoning) is a benchmark for organic multimodal reasoning across mathematics, physics, chemistry, and coding.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- EMMA
- Modality
- multimodal
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- emma|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about EMMA.
Which model scores highest on EMMA?
Seed 2.1 Pro is currently ranked first with 79.3%.
What does EMMA measure?
EMMA (Enhanced MultiModal reAsoning) is a benchmark for organic multimodal reasoning across mathematics, physics, chemistry, and coding.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.