Workflow evaluation for engineering design on Photonics2D
1TCGPT-5-mini
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-5-miniWorkflow Style=W-RAND2026.05 | 1 | 0.53 | |
| GPT-5-miniWorkflow Style=W-DISTRACT2026.05 | 1 | 0.52 | |
| Gemini-3-FlashWorkflow Style=W-RAND2026.05 | 1 | 0.6 | |
| Gemini-3-FlashWorkflow Style=W-DISTRACT2026.05 | 1 | 0.55 | |
| Qwen3-4BWorkflow Style=W-RAND2026.05 | 1 | 0.56 | |
| Qwen3-4BWorkflow Style=W-DISTRACT2026.05 | 1 | 0.52 | |
| Qwen3.5-4BWorkflow Style=W-RAND2026.05 | 1 | 0.55 | |
| Qwen3.5-4BWorkflow Style=W-DISTRACT2026.05 | 1 | 0.57 | |
| Gemini-3-FlashWorkflow Style=W-COND2026.05 | 0.53 | 0.54 | |
| Qwen3.5-4BWorkflow Style=W-COND2026.05 | 0.47 | 0.49 | |
| GPT-5-miniWorkflow Style=W-COND2026.05 | 0.4 | 0.47 | |
| Qwen3-4BWorkflow Style=W-COND2026.05 | 0.2 | 0.43 |