PCB-Bench - QA Placement SBERT: leaderboard
Task-2 open-ended placement question answering scored by SBERT sentence-embedding similarity.
Metric: SBERT similarity (%). Source: digailab.github.io. Saturation forecast: Around January 2028. 9 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Nano | 67.86 | #425 |
| 2 | GPT-5 | 67 | #98 |
| 3 | Gemini 2.5 Flash Lite | 66.62 | #422 |
| 4 | Qwen 3 VL 235B A22B (Thinking) | 66.58 | #238 (Qwen 3 VL 235B A22B) |
| 5 | Gemini 2.5 Pro | 66.37 | #152 |
| 6 | InternVL3-78B | 65.88 | #352 |
| 7 | Qwen 3 VL 8B Instruct | 65.1 | #420 |
| 8 | Llama 4 Maverick | 63.1 | #464 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=pcb-bench-qa-placement-sbert · How It Works · Data refreshed daily, snapshot 2026-10-20.