HELM Lite - MATH Level 1 - Intermediate Algebra: leaderboard
HELM Lite v1.13.0 MATH chain-of-thought run on Level 1 (easiest) Intermediate Algebra problems only: 52 test problems, up to 8-shot, scored by equivalence of the final answer to the reference. Percent, 0-100, higher is better; a small sample, not comparable with full-MATH Intermediate Algebra scores.
Metric: Equivalent (CoT) (%). Source: crfm.stanford.edu. Saturation forecast: Estimated already saturated. 91 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V3 | 86.54 | #329 |
| 2 | Gemini 1.5 Flash (002) | 84.62 | #587 |
| 3 | Qwen 1.5 32B | 84.62 | #794 |
| 4 | Gemini 1.5 Pro (002) | 82.69 | #368 |
| 5 | Claude 3.5 Sonnet (20240620) | 82.69 | #308 |
| 6 | Claude 3.5 Haiku (20241022) | 82.69 | #590 |
| 7 | Nova Pro | 82.69 | #594 |
| 8 | Qwen 2.5 72B Instruct | 80.77 | #469 |
| 9 | Claude 3.5 Sonnet (20241022) | 80.77 | #302 |
| 10 | Mistral Large 2 (Jul) | 80.77 | #490 |
| 11 | GPT-4 Turbo (Preview) | 80.77 | #416 |
| 12 | GPT-4 Turbo | 78.85 | #417 |
| 13 | Gemini 2.0 Flash (Preview) | 78.85 | #376 |
| 14 | Claude 3 Opus (20240229) | 78.85 | #428 |
| 15 | Mistral Large | 75 | #713 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=helm-lite-math-level-1-intermediate-algebra · How It Works · Data refreshed daily, snapshot 2026-11-05.