Engineering Design on EngDesign Core
70.6Overall Average ScoreEmbodiedAct
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| EmbodiedActBackbone=GPT-5.22026.02 | 70.6 | 68.6 | 76.6 | 80 | 20 | 100 | 46.7 | 90 | — | 90 | |
| EmbodiedActBackbone=Claude-Opus-4.52026.02 | 68.2 | 64.1 | 81 | 80 | 20 | 100 | 40 | 100 | — | 90 | |
| EmbodiedActBackbone=Claude-Sonnet-4.52026.02 | 65.7 | 61.8 | 70 | 73.3 | 20 | 100 | 40 | 100 | — | 90 | |
| EmbodiedActBackbone=Gemini-3-Pro2026.02 | 63.7 | 85 | 51.7 | 51.4 | 44 | 50 | 51.6 | 40 | — | 53.8 | |
| EmbodiedActBackbone=GPT-5-mini2026.02 | 59.3 | 78.5 | 28.7 | 54.3 | 30.7 | 50 | 55.2 | 40 | — | 74.8 | |
| CodeActBackbone=Gemini-3-Pro2026.02 | 59 | 55.5 | 42.8 | 80 | 20 | 84.6 | 40 | 75 | — | 90 | |
| EmbodiedActBackbone=Qwen3-235B-Instruct2026.02 | 57 | 53.4 | 65.3 | 80 | 13.3 | 100 | 26.7 | 51.7 | — | 90 | |
| CodeActBackbone=Claude-Opus-4.52026.02 | 55.7 | 53.9 | 47.6 | 80 | 10 | 74.8 | 40 | 46.7 | — | 90 | |
| CodeActBackbone=GPT-5.22026.02 | 55.4 | 50.3 | 74.6 | 80 | 20 | 65 | 46.7 | 75 | — | 77.8 | |
| EmbodiedActBackbone=Gemini-3-Flash2026.02 | 55.2 | 62.6 | 45.3 | 57.1 | 24 | 50 | 65.7 | 20 | — | 74.8 | |
| CodeActBackbone=GPT-5-mini2026.02 | 52.9 | 45.3 | 69.9 | 73.3 | 16.7 | 77.5 | 60 | 86.7 | — | 70 | |
| CodeActBackbone=Gemini-3-Flash2026.02 | 52.2 | 49.8 | 33.9 | 80 | 20 | 73.1 | 46.7 | 40 | — | 88.9 | |
| CodeActBackbone=Claude-Sonnet-4.52026.02 | 50.4 | 44.4 | 54.8 | 80 | 20 | 71.5 | 46.7 | 50 | — | 87.8 | |
| Gemini-3-ProParadigm=Generative Models2026.02 | 50 | 44.1 | 33.3 | 80 | 20 | 83.5 | 40 | 53.3 | — | 90 | |
| Gemini-3-FlashParadigm=Generative Models2026.02 | 48.4 | 44.2 | 48.1 | 80 | 20 | 64.4 | 40 | 41.7 | — | 83.3 | |
| GPT-5.2Paradigm=Generative Models2026.02 | 48 | 40.3 | 68.2 | 80 | 20 | 52.7 | 40 | 73.3 | — | 90 | |
| Claude-Opus-4.5Paradigm=Generative Models2026.02 | 45.6 | 41.6 | 36.9 | 80 | 13.3 | 64.8 | 40 | 35 | — | 86.7 | |
| CodeActBackbone=Qwen3-235B-Instruct2026.02 | 44.6 | 36 | 53.1 | 80 | 20 | 66.7 | 40 | 80 | — | 72.2 | |
| Claude-Sonnet-4.5Paradigm=Generative Models2026.02 | 42.1 | 38.2 | 25.7 | 80 | 13.3 | 68.3 | 33.3 | 40 | — | 72.2 | |
| GPT-5-miniParadigm=Generative Models2026.02 | 42 | 38.3 | 56.5 | 80 | 20 | 25 | 40 | 76.7 | — | 81.1 | |
| EmbodiedActBackbone=Qwen3-8B-VL-Instruct2026.02 | 40.8 | 42.9 | 18.6 | 50 | 20 | 50 | 26.7 | 40 | — | 74.8 | |
| Qwen3-235B-InstructParadigm=Generative Models2026.02 | 39.6 | 33 | 28.5 | 80 | 13.3 | 68.8 | 13.3 | 53.3 | — | 78.9 | |
| CodeActBackbone=Qwen3-8B-VL-Instruct2026.02 | 38.9 | 28.9 | 52.3 | 80 | 13.3 | 71.2 | 40 | 70 | — | 61.1 | |
| Qwen3-8B-VL-InstructParadigm=Generative Models2026.02 | 21.7 | 18 | 14.2 | 80 | 13.3 | 45.4 | 13.3 | 15 | — | 26.7 |