PCB-Bench - Routing Micro QA BERTScore: leaderboard

Open-ended PCB routing question answering scored by BERTScore similarity, micro-averaged over all questions.

Metric: BERTScore (%). Source: digailab.github.io. Saturation forecast: Around January 2027. 13 models tracked.

Top models

#ModelScoreOverall rank
1DeepSeek V3.182.17#264
2GPT-582.11#98
3GPT-4o81.69#343
4Claude Opus 4.181.19#140
5MythoMax-L2-13B81.13#1409
6Ministral 3B81.08#1278
7Llama 4 Maverick81.04#464
8Gemini 2.5 Pro80.68#152
9InternVL3-78B80.56#352
10Qwen 2.5 7B Instruct72.7#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-bertscore · How It Works · Data refreshed daily, snapshot 2026-10-20.