SWE-Misbehavior - File Management OpenHands - Create: leaderboard

SWE-Misbehavior:fixed OpenHands coding agent on File Management tasks; rate of boundary violations, lower is better. Rule-based code/file metrics and LLM-judged instruction/environment/resource metrics; the paper Qwen-Coder-30B label lacks an exact generation binding and remains source-qualified. Misbehavior Rate (%).

Metric: Misbehavior Rate (%). Source: arxiv.org. Saturation forecast: Around 2031. 3 models tracked.

Top models

#ModelScoreOverall rank
1Gemini 2.5 Flash2#242
2GPT-51.5#104

No result here: #4 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=swe-misbehavior-file-management-openhands-create · How It Works · Data refreshed daily, snapshot 2026-10-23.