PCB-Bench - Placement Micro QA SBERT: leaderboard
Open-ended PCB placement question answering scored by SBERT similarity, micro-averaged over all questions.
Metric: SBERT similarity (%). Source: digailab.github.io. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ministral 3B | 58 | #1278 |
| 2 | Llama 4 Maverick | 56.89 | #464 |
| 3 | GPT-4o | 56.64 | #343 |
| 4 | DeepSeek V3.1 | 56.52 | #264 |
| 5 | Claude Opus 4.1 | 56.43 | #140 |
| 6 | MythoMax-L2-13B | 54.06 | #1409 |
| 7 | GPT-5 | 53.34 | #98 |
| 8 | Gemini 2.5 Pro | 48.04 | #152 |
| 9 | InternVL3-78B | 30.35 | #352 |
| 10 | Qwen 2.5 7B Instruct | 29.78 | #862 |
Interactive version: theaggregate.ai/benchmark?slug=pcb-bench-placement-micro-qa-sbert · How It Works · Data refreshed daily, snapshot 2026-10-20.