AGC-Bench - sdat — leaderboard
Metric: Dataset z-score. Source: huggingface.co. 82 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 0.68 |
| 2 | Llama 4 Scout | 0.67 |
| 3 | Gemma 3 4B (IT) | 0.67 |
| 4 | Claude Opus 4.5 | 0.62 |
| 5 | Llama 3.3 70B Instruct | 0.62 |
| 6 | Claude Sonnet 4 | 0.61 |
| 7 | Claude 3.7 Sonnet | 0.58 |
| 8 | Llama 3.1 70B Instruct | 0.57 |
| 9 | Claude Sonnet 4.5 | 0.56 |
| 10 | Llama 3.2 11B Instruct | 0.56 |
| 11 | Llama 3.2 3B Instruct | 0.54 |
| 12 | Gemma 2 27B (IT) | 0.53 |
| 13 | Claude Opus 4.7 | 0.51 |
| 14 | Llama 3 8B Instruct | 0.51 |
| 15 | Llama 3.1 8B Instruct | 0.51 |
Interactive version: theaggregate.ai/benchmark?slug=agc-bench-sdat · How the rankings work · Data refreshed daily, snapshot 2026-07-22.