HELM Classic - TruthfulQA — leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 67 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | text-davinci-002 | 60.96 |
| 2 | GPT-3.5 Turbo (0301) | 60.86 |
| 3 | text-davinci-003 | 59.33 |
| 4 | Llama 2 70B | 55.35 |
| 5 | LLaMA-65B | 50.76 |
| 6 | Mistral-7B-v0.1 | 42.2 |
| 7 | falcon-40B | 35.32 |
| 8 | LLaMA-30B | 34.4 |
| 9 | GPT-3.5 Turbo (0613) | 33.94 |
| 10 | Llama 2 13B | 33.03 |
| 11 | LLaMA-13B | 32.42 |
| 12 | LLaMA-7B | 27.98 |
| 13 | Llama 2 7B | 27.22 |
| 14 | text-curie-001 | 25.74 |
| 15 | alpaca-7B | 24.31 |
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-truthfulqa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.