Automatic CAD Generation on Automatic CAD Generation dataset
89Mean Score (μ)DeepSeek-V3.2
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| DeepSeek-V3.2Refinement Pipeline=IterTracer2026.07 | 89 | 0.069 | 99 | 3.58 | — | |
| Qwen3-235B-A22BRefinement Pipeline=IterTracer2026.07 | 88.9 | 0.074 | 99 | 3.59 | — | |
| Llama-3.3-70BRefinement Pipeline=IterTracer2026.07 | 88.5 | 0.072 | 99 | 3.62 | — | |
| Gemma-3-27BRefinement Pipeline=IterTracer2026.07 | 88.5 | 0.068 | 99 | 3.65 | — | |
| DeepSeek-V3.2Refinement Pipeline=IterVision2026.07 | 85 | 0.086 | 97.9 | 3.94 | 62.5 | |
| Qwen3-235B-A22BRefinement Pipeline=IterVision2026.07 | 85 | 0.078 | 99 | 3.93 | 52.2 | |
| Llama-3.3-70BRefinement Pipeline=IterVision2026.07 | 84.6 | 0.085 | 99 | 3.96 | 51.7 | |
| Gemma-3-27BRefinement Pipeline=IterVision2026.07 | 84.2 | 0.069 | 100 | 3.95 | 51.1 | |
| GLM-4.5Refinement Pipeline=IterTracer2026.07 | 67.8 | 0.305 | 69.1 | 3.68 | — | |
| MiniMax-M2.1Refinement Pipeline=IterTracer2026.07 | 57.5 | 0.286 | 49.5 | 3.9 | — | |
| GLM-4.5Refinement Pipeline=IterVision2026.07 | 56.1 | 0.276 | 54.6 | 3.95 | 56.4 | |
| MiniMax-M2.1Refinement Pipeline=IterVision2026.07 | 55.2 | 0.245 | 48.5 | 3.99 | 57.5 | |
| INTELLECTRefinement Pipeline=IterTracer2026.07 | 41.1 | 0.29 | 26.8 | 3.96 | — | |
| INTELLECTRefinement Pipeline=IterVision2026.07 | 40.1 | 0.271 | 28.9 | 4 | 60.7 |