New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
MATH-500 benchmark
AI model leaderboard for the MATH-500 benchmark. Compare how large language models score on MATH-500, see the full ranking, and understand what this AI benchmark measures. DeepSeek-R1-0528 currently leads with 99.3%. 500-problem subset of MATH curated by OpenAI for benchmark stability.
Leaderboard
Frequently asked questions about MATH-500
What is the MATH-500 benchmark?
500-problem subset of MATH curated by OpenAI for benchmark stability.
How is the MATH-500 benchmark scored?
MATH-500 is scored using the accuracy (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on MATH-500?
As of the latest reported scores on GenAIList, DeepSeek-R1-0528 achieves the highest result on MATH-500 with a score of 99.3%.
Is a higher MATH-500 score better?
Yes. On MATH-500 a higher score indicates better performance, so models near the top of the leaderboard are the strongest.