CRMArena - KQA (Function Calling Native): leaderboard
Metric: F1 (%). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude 3.5 Sonnet (20240620) | 40.5 |
| 2 | GPT-4o Mini | 39.7 |
| 3 | GPT-4o (2024-08-06) | 30.4 |
| 4 | Claude 3 Sonnet (20240229) | 23.2 |
Interactive version: theaggregate.ai/benchmark?slug=crmarena-kqa-function-calling-native · How It Works · Data refreshed daily, snapshot 2026-10-09.