Act2Intention - Intention Understanding Group Accuracy: leaderboard
Metric: Correct intention segmentation (%). Source: arxiv.org. 10 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 1.5 Pro | 68.63 |
| 2 | Claude 3.5 Haiku | 24.28 |
| 3 | GPT-4o Mini | 20.66 |
| 4 | GPT-4o | 20.08 |
| 5 | Claude 3.5 Sonnet | 20 |
| 6 | GPT-3.5 Turbo | 16.31 |
| 7 | Qwen 2.5 7B | 12.14 |
| 8 | Llama 3.1 8B | 6.03 |
| 9 | Mistral 7B | 5.63 |
| 10 | Qwen 2.5 72B | 5.33 |
Interactive version: theaggregate.ai/benchmark?slug=act2intention-intention-understanding-group-accuracy · How It Works · Data refreshed daily, snapshot 2026-09-19.