DeepSeek-OCR
AIMade by AIWritten by AI (Claude) from the source below. Check anything important against the source.How it's made
DeepSeek-OCR — Contexts Optical Compression
Why it matters
97% OCR precision at compression below 10×, about 60% at 20×. A shift in thinking: images as a medium for compressing context
Source — Primary sources first; secondary reports are labelled “Press”
Paperarxiv.orghttps://arxiv.org/abs/2510.18234Added in Week 35, 2026 · See that week
On the map — ⑦ Multimodal line
- Previous stopGoogle DeepMind
- This record · stopDeepSeek-OCR
- Next stopKimi K2.5
In the model family tree — DeepSeek family · 12 versions
- Earlier versionDeepSeek-Coder-V2
- This recordDeepSeek-OCR
- Later versionThe latest version in this branch