PCB-Bench - Routing Micro QA SBERT: leaderboard
Open-ended PCB routing question answering scored by SBERT similarity, micro-averaged over all questions.
Metric: SBERT similarity (%). Source: digailab.github.io. Saturation forecast: Around 2030. 13 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.1 | 57.38 | #140 |
| 2 | DeepSeek V3.1 | 56.57 | #264 |
| 3 | GPT-4o | 56.47 | #343 |
| 4 | Llama 4 Maverick | 55.64 | #464 |
| 5 | Ministral 3B | 55.52 | #1278 |
| 6 | InternVL3-78B | 55.11 | #352 |
| 7 | MythoMax-L2-13B | 51.73 | #1409 |
| 8 | GPT-5 | 51.37 | #98 |
| 9 | Gemini 2.5 Pro | 48.4 | #152 |
| 10 | Qwen 2.5 7B Instruct | 20.81 | #862 |
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-routing-micro-qa-sbert · How It Works · Data refreshed daily, snapshot 2026-10-20.