DAREBench - Multimodal - Single-Step: leaderboard
Metric: Accuracy (%). Source: arxiv.org. 27 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 90 |
| 2 | Gemini 3.1 Pro (Preview) | 85 |
| 3 | Claude Opus 4.8 | 85 |
| 4 | Gemini 3.5 Flash | 85 |
| 5 | GPT-5.4 | 82.5 |
| 6 | MiMo-V2.5 | 80 |
| 7 | Claude Sonnet 4.6 | 75 |
| 8 | Claude Opus 4.6 | 75 |
| 9 | Qwen 3.6 Plus | 75 |
| 10 | Qwen 3.6 27B | 75 |
| 11 | MiMo-V2-Omni | 75 |
| 12 | GPT-5.4 Mini | 70 |
| 13 | Qwen 3.5 Plus | 70 |
| 14 | Kimi K2.5 | 65 |
| 15 | Qwen 3.6 35B A3B | 65 |
Interactive version: theaggregate.ai/benchmark?slug=darebench-multimodal-single-step · How It Works · Data refreshed daily, snapshot 2026-09-19.