MUSE (Text-to-CAD) - Code Execution: leaderboard
Metric: Sandbox success (%): share of the 106 design specifications whose generated CadQuery script runs in the sandbox and exports a STEP model, one CadQuery script per design specification, executed in a sandbox; funnel evaluation, so a sample that fails a stage scores zero on every later stage; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 15 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 77.36 |
| 2 | Claude Opus 4.7 | 76.42 |
| 3 | Gemini 3.1 Pro (Preview) | 65.09 |
| 4 | Qwen 2.5 72B | 50 |
| 5 | Claude 3.7 Sonnet | 46.23 |
| 6 | GPT-4o | 36.79 |
| 7 | GLM-5.1 | 31.13 |
| 8 | Llama 3.1 70B | 31.13 |
| 9 | Qwen 3.5 122B A10B | 27.36 |
| 10 | MiniMax-M2.5 | 25.47 |
| 11 | MiniMax-M2.7 | 18.87 |
| 12 | Qwen 3.6 35B A3B | 8.49 |
| 13 | GLM-4.7 Flash | 8.49 |
| 14 | Llama 3.1 8B | 2.83 |
Interactive version: theaggregate.ai/benchmark?slug=muse-text-to-cad-code-execution · How It Works · Data refreshed daily, snapshot 2026-10-07.