PCB-Bench - Routing Macro QA BERTScore: leaderboard

Open-ended question answering about PCB routing, scored by BERTScore similarity and macro-averaged over boards.

Metric: BERTScore (%). Source: digailab.github.io. Saturation forecast: Not forecast. 13 models tracked.

Top models

#ModelScoreOverall rank
1DeepSeek V3.183.1#264
2GPT-4o82.67#343
3MythoMax-L2-13B82.62#1409
4Ministral 3B82.04#1278
5GPT-581.93#98
6Llama 4 Maverick81.86#464
7Claude Opus 4.181.64#140
8InternVL3-78B80.71#352
9Gemini 2.5 Pro80.68#152
10Qwen 2.5 7B Instruct73.05#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-macro-qa-bertscore · How It Works · Data refreshed daily, snapshot 2026-10-20.