BlueBench - Entity Extraction: leaderboard
Metric: Score (%). Source: huggingface.co. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O1 | 79.22 |
| 2 | O4 Mini | 75.47 |
| 3 | Mistral Medium 3 | 74.21 |
| 4 | GPT-4o | 73.62 |
| 5 | GPT-4.1 Mini | 73.17 |
| 6 | O3 Mini | 72.86 |
| 7 | GPT-4.1 | 72.39 |
| 8 | Llama 3.3 70B Instruct | 70.18 |
| 9 | Mistral Large | 61.18 |
| 10 | GPT-4.1 Nano | 61.11 |
| 11 | Llama 3.2 3B Instruct | 45.26 |
| 12 | Granite 3.3 8B Instruct | 44.27 |
| 13 | Pixtral-12B | 22 |
| 14 | Llama 3.2 1B Instruct | 20.69 |
Interactive version: theaggregate.ai/benchmark?slug=bluebench-entity-extraction · How It Works · Data refreshed daily, snapshot 2026-09-05.