CRMArena - NCR (ReAct): leaderboard

Metric: Exact match (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 9 models tracked.

Top models

#ModelScore
1Llama 3.1 405B Instruct81.5
2GPT-4o (2024-08-06)70
3O1 (2024-12-17)70
4Claude 3.5 Sonnet (20240620)62.9
5DeepSeek R153.8
6Llama 3.1 70B Instruct48.5
7GPT-4o Mini40.8
8Claude 3 Sonnet (20240229)7.7
9Llama 3.1 8B Instruct0

Interactive version: theaggregate.ai/benchmark?slug=crmarena-ncr-react · How It Works · Data refreshed daily, snapshot 2026-10-09.