Natural Questions
AI benchmark results and model performance
Type: percentageGeneral
Recorded results
3
Average score
22.53%
Score range
15.50% - 31.20%
Leading model
31.20% - Mistral Nemo 12B
Scores over time
Individual benchmark scores plotted by date.
Model results
| Organisation | Models | Reported | Top score | Info | Self reported | Source |
|---|---|---|---|---|---|---|
| Mistral Nemo 12B | 18 Jul 2024 | 31.20% | inferred family alias from mistral-nemo-instruct-2407 (score=0.3250; benches=8) | Yes | Source | |
| Gemma 3n E4B | 25 Jun 2025 | 20.90% | - | Yes | Source | |
| Gemma 3n E2B (Free) | 25 Jun 2025 | 15.50% | - | Yes | Source |
