AGC-Bench - science_analogies — leaderboard
Metric: Dataset z-score. Source: huggingface.co. 81 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 2.5 |
| 2 | Hermes 4 405B | 2.45 |
| 3 | GPT-4o Mini | 2.15 |
| 4 | Hermes 4 70B | 2 |
| 5 | trinity-large-preview | 1.94 |
| 6 | Hermes 3 - Llama-3.1 70B | 1.5 |
| 7 | GLM-5.1 | 1.15 |
| 8 | Qwen 3 14B | 1.03 |
| 9 | jamba-large-1.7 | 0.89 |
| 10 | Gemma 4 31B (IT) | 0.79 |
| 11 | ERNIE 4.5 300B A47B | 0.63 |
| 12 | GPT-5.5 | 0.57 |
| 13 | GLM-5 | 0.53 |
| 14 | Mistral Large 3 | 0.52 |
| 15 | Nova Pro (v1) | 0.52 |
Interactive version: theaggregate.ai/benchmark?slug=agc-bench-science-analogies · How the rankings work · Data refreshed daily, snapshot 2026-07-22.