PCB-Bench - Task 3 Recall — leaderboard
Metric: Recall (%). Source: digailab.github.io. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3 VL 8B Instruct | 89.55 |
| 2 | InternVL3-78B | 87.16 |
| 3 | GPT-5 | 86.91 |
| 4 | Gemini 2.5 Flash Lite | 85.18 |
| 5 | Qwen 3 VL 235B A22B (Thinking) | 83.91 |
| 6 | GPT-5 Nano | 83.52 |
| 7 | Llama 4 Maverick | 80.44 |
| 8 | Gemini 2.5 Pro | 75.07 |
Interactive version: theaggregate.ai/benchmark?slug=pcb-bench-task-3-recall · How the rankings work · Data refreshed daily, snapshot 2026-07-22.