FrontierCode 2.0
Cognition's second-generation coding-agent benchmark, scoring whether agents produce mergeable pull requests on larger, less fully specified tasks, with added emphasis on thoroughness such as fixing every instance of a bug.
ModelsModels with a result.27
Top-3 spreadPoints from first to third.6.3 pts
YearYear of release.—
Result and price
24 of 27 results
Results
27 results#ModelResultThe score from the source.
