HELM Classic - bAbI — leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 70 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 2 70B | 70.53 |
| 2 | code-davinci-002 | 68.62 |
| 3 | LLaMA-65B | 67.66 |
| 4 | text-davinci-003 | 65.34 |
| 5 | LLaMA-30B | 63.36 |
| 6 | text-davinci-002 | 61.81 |
| 7 | Mistral-7B-v0.1 | 59.27 |
| 8 | Llama 2 13B | 58.39 |
| 9 | falcon-40B | 57.62 |
| 10 | LLaMA-13B | 56.07 |
| 11 | mpt-30B | 54.97 |
| 12 | Llama 2 7B | 54.97 |
| 13 | LLaMA-7B | 53.09 |
| 14 | GPT-3.5 Turbo (0613) | 52.76 |
| 15 | GPT-3.5 Turbo (0301) | 51.55 |
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-babi · How the rankings work · Data refreshed daily, snapshot 2026-07-22.