HELM Classic - NaturalQuestions Open Book — leaderboard
Metric: F1 (%). Source: crfm.stanford.edu. 66 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | text-davinci-003 | 77.02 |
| 2 | text-davinci-002 | 71.32 |
| 3 | Mistral-7B-v0.1 | 68.66 |
| 4 | falcon-40B | 67.53 |
| 5 | GPT-3.5 Turbo (0613) | 67.48 |
| 6 | Llama 2 70B | 67.42 |
| 7 | mpt-30B | 67.29 |
| 8 | LLaMA-65B | 67.21 |
| 9 | LLaMA-30B | 66.56 |
| 10 | Llama 2 13B | 63.73 |
| 11 | davinci | 62.46 |
| 12 | GPT-3.5 Turbo (0301) | 62.43 |
| 13 | LLaMA-13B | 61.43 |
| 14 | Llama 2 7B | 61.13 |
| 15 | gpt-neox-20B | 59.61 |
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-naturalquestions-open-book · How the rankings work · Data refreshed daily, snapshot 2026-07-22.