math benchmark
FunctionalMATH
A functional variant of the MATH benchmark that tests language models' ability to generalize reasoning patterns across different problem instances, revealing the reasoning gap between static and functional performance.
Updated Aug 11, 2026
Models2
Model coverage2
MetricScore
EvidenceB
FunctionalMATH Ranking
Higher score ranks better on this benchmark.
| 1 | GO | 64.6% | 100.0% | 2 | C | |
| 2 | GO | 53.6% | 0.0% | 2 | C |
FunctionalMATH Score Distribution
A closer view of the leading scores on this benchmark.
FunctionalMATH Highlights
The leading models and scores on this benchmark.
What is FunctionalMATH?
What FunctionalMATH measures and how its scores work.
A functional variant of the MATH benchmark that tests language models' ability to generalize reasoning patterns across different problem instances, revealing the reasoning gap between static and functional performance.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- FunctionalMATH
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- functionalmath|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about FunctionalMATH.
Which model scores highest on FunctionalMATH?
Gemini 1.5 Pro is currently ranked first with 64.6%.
What does FunctionalMATH measure?
A functional variant of the MATH benchmark that tests language models' ability to generalize reasoning patterns across different problem instances, revealing the reasoning gap between static and functional performance.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 model results are currently shown.
Does this benchmark affect the overall score?
No. This benchmark is shown for reference but does not contribute to the overall score.