CRMArena - HTU — leaderboard
Metric: HTU Score (%). Source: huggingface.co. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 | 68.5 |
| 2 | GPT-4o | 47.7 |
| 3 | GPT-4o Mini | 38.5 |
| 4 | Claude 3.5 Sonnet | 33.1 |
| 5 | Llama 3.1 405B | 31.5 |
| 6 | Claude 3 Sonnet | 26.9 |
| 7 | Llama 3.1 70B | 23.1 |
| 8 | DeepSeek R1 | 23.1 |
| 9 | Llama 3.1 8B | 0 |
Interactive version: theaggregate.ai/benchmark?slug=crmarena-htu · How the rankings work · Data refreshed daily, snapshot 2026-07-22.