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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Atlassian Rovo Dev (2025-12-08) | 46.38 | |
| 2 | iSWE-Agent | 46.38 | |
| 3 | HMigBot | 43.48 | |
| 4 | Prometheus-v1.2 + GPT-5 | 33.33 | |
| 5 | Kodah | 27.54 |
Interactive version: theaggregate.ai/benchmark?slug=swe-polybench-verified-java · How It Works · Data refreshed daily, snapshot 2026-10-20.