AGC-Bench - creatset — leaderboard
Metric: Dataset z-score. Source: huggingface.co. 83 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.2 3B Instruct | 1.63 |
| 2 | Mistral Small 3.1 | 1.61 |
| 3 | Hermes 3 - Llama-3.1 70B | 1.61 |
| 4 | Llama 3 8B Instruct | 1.48 |
| 5 | Llama 3.2 11B Instruct | 1.47 |
| 6 | Llama 3.1 70B Instruct | 1.43 |
| 7 | Llama 3.1 8B Instruct | 1.37 |
| 8 | granite-4.0-h-micro | 1.37 |
| 9 | Hermes 4 405B | 1.37 |
| 10 | Mistral Nemo | 1.3 |
| 11 | Llama 4 Scout | 1.24 |
| 12 | Hermes 4 70B | 1.22 |
| 13 | Llama 3.3 70B Instruct | 1.18 |
| 14 | Gemma 2 27B (IT) | 1.15 |
| 15 | Llama 4 Maverick | 1.13 |
Interactive version: theaggregate.ai/benchmark?slug=agc-bench-creatset · How the rankings work · Data refreshed daily, snapshot 2026-07-22.