CAD Generation on Hybrid Agentic Architecture CAD Evaluation Set
97.5Execution Success (R1)Claude Sonnet 4.5
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Claude Sonnet 4.5Model=Claude Sonnet 4.52026.05 | 97.5 | 87.6 | 84.5 | |
| Gemini 3 ProModel=Gemini 3 Pro2026.05 | 97.5 | 76.3 | 86.8 | |
| Gemini 3 FlashModel=Gemini 3 Flash2026.05 | 97 | 84.4 | 85.1 | |
| Claude Opus 4.5Model=Claude Opus 4.52026.05 | 96.4 | 88.2 | 81.3 |