BLINKG - Basic - Entity Class: leaderboard
Metric: F1 (%; expert-validated mappings for the ontology class of each generated entity; 12 gold task instances; eight atomic CSV cases over a three-class ontology; macro-average of three runs). Source: arxiv.org. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Pro | 97 |
| 2 | GPT-4o | 91 |
| 3 | Llama 3.3 70B Instruct | 91 |
| 4 | O3 | 90 |
| 5 | Mixtral 8x22B Instruct | 88 |
Interactive version: theaggregate.ai/benchmark?slug=blinkg-basic-entity-class · How It Works · Data refreshed daily, snapshot 2026-09-26.