AGC-Bench - fann_or_flop — leaderboard
Metric: Dataset z-score. Source: huggingface.co. 83 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | trinity-large-preview | 2.3 |
| 2 | Llama 4 Maverick | 1.74 |
| 3 | GLM-5.1 | 1.37 |
| 4 | ERNIE 4.5 300B A47B | 1.36 |
| 5 | GPT-5.1 | 1.27 |
| 6 | Kimi K2 0905 | 1.16 |
| 7 | GLM-4.7 | 1.14 |
| 8 | Claude 3.7 Sonnet | 1.13 |
| 9 | GPT-4o | 1.04 |
| 10 | Claude Sonnet 4.5 | 1.02 |
| 11 | DeepSeek V3 Chat | 0.92 |
| 12 | GLM-5-Turbo | 0.83 |
| 13 | DeepSeek V3 (0324) | 0.77 |
| 14 | Hermes 4 405B | 0.74 |
| 15 | DeepSeek V3.2 | 0.68 |
Interactive version: theaggregate.ai/benchmark?slug=agc-bench-fann-or-flop · How the rankings work · Data refreshed daily, snapshot 2026-07-22.