Build-A-Chip-Bench GDS Pass
Agents · 2026-10-14
Share of all Build-A-Chip-Bench tasks whose design completes the physical flow to a GDS layout that routes, has no hold-time violations and is DRC-clean. It is one of the two pass/fail gates inside the headline score, so it is kept as a diagnostic and never enters capability.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 85.0 |
| Claude Fable 5.1 | 75.0 |
| Gemini 3.8 Flash | 45.0 |
| Kimi K3 | 25.0 |
| GLM-5.3 | 5.0 |