PCB-Bench - Placement Micro QA BERTScore: leaderboard
Open-ended PCB placement 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
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 | 82.14 | #98 |
| 2 | DeepSeek V3.1 | 81.27 | #264 |
| 3 | MythoMax-L2-13B | 81.01 | #1409 |
| 4 | Llama 4 Maverick | 80.74 | #464 |
| 5 | GPT-4o | 80.69 | #343 |
| 6 | Gemini 2.5 Pro | 80.69 | #152 |
| 7 | Claude Opus 4.1 | 80.54 | #140 |
| 8 | Ministral 3B | 80.27 | #1278 |
| 9 | InternVL3-78B | 73.6 | #352 |
| 10 | Qwen 2.5 7B Instruct | 72.87 | #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-placement-micro-qa-bertscore · How It Works · Data refreshed daily, snapshot 2026-10-20.