LLM Stats (SWE-Bench Multimodal): leaderboard
LLM Stats compilation of vendor-reported SWE-bench Multimodal results: GitHub issues from JavaScript front-end and visualisation libraries whose reports carry screenshots, mockups or diagrams (517-task test split), resolved when the model's patch passes the repository tests. Percent resolved, higher is better. All rows are self-reported and harnesses differ by vendor (Anthropic reports its internal harness), so this is not the swebench.com board.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around September 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 61.4 | #3 |
| 2 | Claude Mythos Preview | 59 | #7 |
| 3 | Claude Sonnet 5.5 | 54.3 | #13 |
| 4 | Qwen 3.8 27B | 38.6 | #125 |
| 5 | Claude Opus 4.8 | 38.4 | #42 |
| 6 | Claude Haiku 5.5 | 30.7 | #46 |
| 7 | Claude Sonnet 5 | 28.1 | #67 |
No result here: #5 GPT-6 Astra, #6 GPT-6.1 Sol, #9 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-swe-bench-multimodal · How It Works · Data refreshed daily, snapshot 2026-11-07.