IndustryCode - MATLAB Sub-Problems: leaderboard
Metric: Pass@1 (%) on the MATLAB sub-problems of the IndustryCode test split (106 main problems decomposed into 499 sub-problems; industrial Python, C++, MATLAB and Stata projects with test cases), each sub-problem generated with the main-problem description and the code of earlier steps as context; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 77.9 |
| 2 | Claude Sonnet 4.5 | 74.3 |
| 3 | Claude Opus 4.5 | 72.8 |
| 4 | Claude Haiku 4.5 | 64.3 |
| 5 | Gemini 2.5 Pro | 60 |
| 6 | Grok 4.1 | 54.2 |
| 7 | Kimi K2 | 50.7 |
| 8 | Qwen 3 Max (Thinking) | 49.3 |
| 9 | GPT-5.2 | 47.1 |
| 10 | GPT-5.1 | 47.1 |
| 11 | Qwen 3 Max | 46.4 |
| 12 | GLM-4.7 (Non-reasoning) | 45 |
| 13 | GLM-4.7 (Thinking) | 41.4 |
| 14 | Kimi K2 (Thinking) | 40 |
| 15 | Qwen 3 Coder Plus | 40 |
Interactive version: theaggregate.ai/benchmark?slug=industrycode-matlab-sub-problems · How It Works · Data refreshed daily, snapshot 2026-10-07.