Written by Jakub Rusinowski · Last updated September 11, 2026
A 3B vision-language model built for one job — turning document pages into structured text — and a demonstration of "contexts optical compression": rendering text as an image and decoding it costs roughly 10x fewer tokens at near-lossless fidelity, or 20x at about 60% precision. A DeepEncoder compresses each page into vision tokens which a DeepSeek-3B MoE decoder reads, activating only ~570M parameters. A second generation, DeepSeek-OCR-2, followed in January 2026.
| Licence | What it permits | Applies to |
|---|---|---|
MIT | Commercial use permitted Commercial use permitted. No usage restrictions beyond attribution. | DeepSeek-OCR 3B |
| DeepSeek-OCR 3B | Min 3 GB VRAM · Q4_K_M · 8,192 ctx · |
The cheapest GPU that runs DeepSeek-OCR locally (min 3 GB VRAM) is the Intel Arc B570 (10 GB).
Install Ollama then run: ollama run
Minimum VRAM: 3 GB. For best results use Q4_K_M quantization.
DeepSeek-OCR needs about 3 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: DeepSeek-OCR 3B (3 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Yes — DeepSeek-OCR runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
Q4_K_M is the best balance of quality and VRAM for DeepSeek-OCR in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
Install Ollama, then run: ollama run . This downloads DeepSeek-OCR and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.