PCB-Bench - QA Routing SBERT — leaderboard
Metric: SBERT similarity (%). Source: digailab.github.io. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5 Nano | 63.47 |
| 2 | Gemini 2.5 Flash Lite | 63.12 |
| 3 | Gemini 2.5 Pro | 63.11 |
| 4 | InternVL3-78B | 62.42 |
| 5 | Llama 4 Maverick | 61.08 |
| 6 | GPT-5 | 60.55 |
| 7 | Qwen 3 VL 235B A22B (Thinking) | 60.37 |
| 8 | Qwen 3 VL 8B Instruct | 59.97 |
Interactive version: theaggregate.ai/benchmark?slug=pcb-bench-qa-routing-sbert · How the rankings work · Data refreshed daily, snapshot 2026-07-22.