CRMArena - PVI (ReAct): leaderboard
Metric: Exact match (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.1 70B Instruct | 37.7 |
| 2 | GPT-4o (2024-08-06) | 35.4 |
| 3 | DeepSeek R1 | 34.6 |
| 4 | Llama 3.1 405B Instruct | 34.6 |
| 5 | Claude 3.5 Sonnet (20240620) | 30 |
| 6 | O1 (2024-12-17) | 30 |
| 7 | Claude 3 Sonnet (20240229) | 28.5 |
| 8 | GPT-4o Mini | 24.6 |
| 9 | Llama 3.1 8B Instruct | 15.4 |
Interactive version: theaggregate.ai/benchmark?slug=crmarena-pvi-react · How It Works · Data refreshed daily, snapshot 2026-10-09.