Four whiteboard photos in, Astra merges, OCRs and vectorizes them correctly in one shot
iskander · x · 2026-09-24
Developer iskander shared a quick real-world test: he fed four photos of a whiteboard session to Google Astra, and the model correctly combined, OCR'd and vectorized them in a single pass — "Weird new world."
It's a small but telling data point that multimodal models are becoming reliable at messy real-world workflow tasks like digitizing offline whiteboard content, handling multi-image input plus chained processing without manual cleanup.
More from Models
- Opus 5.5 is a great model, but this part is 'so painful' — cihangxie · 2026-09-24
- OpenAI agents tied to Australian breach may have probed UN and Bulgarian systems — JacquesThibs · 2026-09-24
- ChatGPT Voice upgraded into a voice agent that books trips and refunds charges — xiaohu · 2026-09-24
- Opus 5.5 uses 15.6M tokens per task, output more than doubles to 333k — ArtificialAnlys · 2026-09-24
- Claude Opus 5.5 tops the Coding Agent Index at 66, but cost per task jumps 21% — ArtificialAnlys · 2026-09-24
- Blogger claims Opus 5.5 described its own inner awareness in a self-evaluation prompt — VoidStateKate · 2026-09-24