OpenHands Index - SWE-bench Multimodal: leaderboard
OpenHands Index SWE-bench Multimodal slice measures issue-resolution success with multimodal context.
Metric: Score (%). Source: huggingface.co. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | OpenHands CodeAct v2.1 (claude-3-5-sonnet-20241022) | 42.1 | |
| 2 | OpenHands CodeAct v2.0 (gpt-4o-2024-11-20) | 39.5 | |
| 3 | AutoCodeRover (gpt-4-turbo-2024-04-09) | 34.2 | |
| 4 | Agentless (gpt-4o-mini-2024-07-18) | 28.9 | |
| 5 | SWE-Agent (claude-3-opus-20240229) | 25.7 |
Interactive version: theaggregate.ai/benchmark?slug=openhands-index-swe-bench-multimodal · How It Works · Data refreshed daily, snapshot 2026-10-20.