math benchmark
DynaMath
A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.
Updated Aug 7, 2026
Published models7
Registry coverage7
MetricScore
EvidenceB
DynaMath leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 88.0% | 100.0% | 7 | C | |
| 2 | AC | 87.7% | 83.3% | 7 | C | |
| 3 | AC | 85.9% | 66.7% | 7 | C | |
| 4 | AC | 85.6% | 50.0% | 7 | C | |
| 5 | AC | 85.0% | 33.3% | 7 | C | |
| 6 | BY | 73.1% | 16.7% | 7 | C | |
| 7 | BY | 68.1% | 0.0% | 7 | C |
Score distribution
Top published rows on the benchmark's original scale.
DynaMath highlights
The top published results on this benchmark's own scale.
What is DynaMath?
Definition and scoring fields from the benchmark registry.
A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- DynaMath
- Modality
- multimodal
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- dynamath|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about DynaMath.
Which model scores highest on DynaMath?
Qwen3.6 Plus is currently ranked first with 88.0%.
What does DynaMath measure?
A multimodal mathematics and reasoning benchmark focused on dynamic visual problem solving.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
7 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.