AttuneBench — leaderboard
Metric: Composite (self-reported). Source: benchmarklist.com. 11 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 54.3 |
| 2 | GPT-5.5 | 53.71 |
| 3 | Claude Opus 4.7 | 53.64 |
| 4 | MiMo-V2-Pro | 53.61 |
| 5 | Gemini 3.1 Pro (Preview) | 52.9 |
| 6 | Claude Haiku 4.5 | 52.83 |
| 7 | Qwen 2.5 72B Instruct | 52.23 |
| 8 | Claude Sonnet 4.6 | 50.29 |
| 9 | GPT-5.4 | 50.24 |
| 10 | Grok 4 | 50.14 |
Interactive version: theaggregate.ai/benchmark?slug=attunebench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.