DAREBench - Text - Single-Step: leaderboard
Metric: Accuracy (%). Source: arxiv.org. 35 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.8 | 98.8 |
| 2 | Gemini 3.1 Pro (Preview) | 96.4 |
| 3 | Qwen 3.6 Plus | 95.5 |
| 4 | Gemini 3.5 Flash | 94.9 |
| 5 | MiMo-V2.5 | 94.7 |
| 6 | GPT-5.5 | 94.1 |
| 7 | GPT-5.4 | 92.1 |
| 8 | MiMo-V2-Pro | 91.7 |
| 9 | DeepSeek V4 Pro | 91.6 |
| 10 | MiMo-V2.5-Pro | 91.6 |
| 11 | Claude Opus 4.6 | 91.5 |
| 12 | Qwen 3.5 Plus | 89.5 |
| 13 | Claude Sonnet 4.6 | 89.2 |
| 14 | Kimi K2.5 | 89.2 |
| 15 | Claude Haiku 4.5 | 89.1 |
Interactive version: theaggregate.ai/benchmark?slug=darebench-text-single-step · How It Works · Data refreshed daily, snapshot 2026-09-19.