HELM Classic - LSAT: leaderboard
The "LSAT" subtask of HELM Classic. A scenario from Stanford HELM Classic, the holistic evaluation suite scoring language models across accuracy, robustness and calibration on a broad set of tasks.
Metric: Exact Match (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 69 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | mpt-30B | 24.78 | #1476 |
| 2 | RedPajama-INCITE-7B-Instruct | 24.35 | #1566 |
| 3 | Llama 2 13B | 23.48 | #1436 |
| 4 | Llama 2 70B | 23.48 | #1201 |
| 5 | text-davinci-003 | 23.33 | #1146 |
| 6 | text-davinci-002 | 22.9 | #1225 |
| 7 | Palmyra X (43B) | 22.46 | #1317 |
| 8 | alpaca-7B | 22.17 | #1611 |
| 9 | OPT-175B | 22.03 | #1538 |
| 10 | Mistral-7B-v0.1 | 21.3 | #1326 |
| 11 | LLaMA-65B | 21.3 | #1303 |
| 12 | mpt-30B Instruct | 21.3 | #1470 |
| 13 | RedPajama-INCITE-Base-3B-v1 | 21.3 | #1618 |
| 14 | falcon-40B | 21.3 | #1442 |
| 15 | LLaMA-30B | 21.3 | #1423 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-lsat · How It Works · Data refreshed daily, snapshot 2026-10-20.