HELM Classic - LSAT — leaderboard
Metric: Exact Match (%). Source: crfm.stanford.edu. 69 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-3.5 Turbo (0301) | 25.22 |
| 2 | mpt-30B | 24.78 |
| 3 | Llama 2 13B | 23.48 |
| 4 | Llama 2 70B | 23.48 |
| 5 | text-davinci-003 | 23.33 |
| 6 | text-davinci-002 | 22.9 |
| 7 | alpaca-7B | 22.17 |
| 8 | Mistral-7B-v0.1 | 21.3 |
| 9 | LLaMA-65B | 21.3 |
| 10 | LLaMA-30B | 21.3 |
| 11 | falcon-40B | 21.3 |
| 12 | text-ada-001 | 21.3 |
| 13 | LLaMA-7B | 20.87 |
| 14 | falcon-7B | 20.43 |
| 15 | pythia-6.9B | 19.57 |
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-lsat · How the rankings work · Data refreshed daily, snapshot 2026-07-22.