HELM Classic - MATH Chain-of-Thought — leaderboard
Metric: Equivalent (%). Source: crfm.stanford.edu. 69 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-3.5 Turbo (0613) | 71.86 |
| 2 | GPT-3.5 Turbo (0301) | 68.94 |
| 3 | text-davinci-003 | 44.92 |
| 4 | code-davinci-002 | 43.33 |
| 5 | text-davinci-002 | 38.06 |
| 6 | Llama 2 70B | 33 |
| 7 | Mistral-7B-v0.1 | 29.34 |
| 8 | LLaMA-65B | 25.25 |
| 9 | LLaMA-30B | 22.29 |
| 10 | falcon-40B | 13.01 |
| 11 | mpt-30B | 12.41 |
| 12 | Llama 2 13B | 12.18 |
| 13 | LLaMA-13B | 10.98 |
| 14 | Llama 2 7B | 9.91 |
| 15 | gpt-neox-20B | 7.06 |
Interactive version: theaggregate.ai/benchmark?slug=helm-classic-math-chain-of-thought · How the rankings work · Data refreshed daily, snapshot 2026-07-22.