SWE-PolyBench Verified (Java): leaderboard

SWE-PolyBench Verified Java slice evaluates agents on resolving Java GitHub issues from the verified set.

Metric: Score (%). Source: huggingface.co. 5 models tracked.

Top models

#ModelScoreOverall rank
1Atlassian Rovo Dev (2025-12-08)46.38
2iSWE-Agent46.38
3HMigBot43.48
4Prometheus-v1.2 + GPT-533.33
5Kodah27.54

Interactive version: theaggregate.ai/benchmark?slug=swe-polybench-verified-java · How It Works · Data refreshed daily, snapshot 2026-10-20.