SWE-Bench Pro (Public)
Score distribution
Where models cluster
Correlated benchmarks
Pearson r · original research
Frequently asked
About SWE-Bench Pro (Public)
What does SWE-Bench Pro (Public) measure?
SWE-Bench Pro (Public) is a knowledge benchmark in the BenchGecko catalog. 2 AI models have been tested on it. Scores range from 41.0 to 45.9 out of 100.
Which model leads on SWE-Bench Pro (Public)?
Claude Opus 4.5 from Anthropic leads SWE-Bench Pro (Public) with a score of 45.9. The median score across 2 tested models is 43.5.
Is SWE-Bench Pro (Public) saturated?
No · the top score is 45.9 out of 100 (46%). There is still meaningful room for improvement on SWE-Bench Pro (Public).
What makes SWE-Bench Pro (Public) distinctive?
SWE-Bench Pro (Public) is a knowledge benchmark with limited overlap to the rest of the catalog · it measures capabilities that are not well-covered by other benchmarks we track.
How often is SWE-Bench Pro (Public) data refreshed?
BenchGecko pulls updates daily. New model scores on SWE-Bench Pro (Public) appear as soon as they are published by Epoch AI or the model provider.
- Category
- Knowledge
- Max score
- 100
- Models
- 2
- Updated
- 2026-01-14
More knowledge benchmarks
Same category · related evaluations