deepseek-ai / image-text-to-text updated 9 months ago

DeepSeek-OCR

Name: DeepSeek-OCR
Author: deepseek-ai

🌟 Github 📥 Model Download 📄 Paper Link 📄 Arxiv Paper Link DeepSeek-OCR: Contexts Optical Compression Explore the boundaries of visual-text compression.

Params

3.3 B

Context

8,192

Downloads 30d

2.3 M

Likes

3,333

Commercial use: allowed mit Not gated SAFETENSORS 1 languages View on Hugging Face ↗

Download history

daily snapshots · 10 days

2.5 M2.3 M

Jul 28Jul 31Aug 3Aug 6

Can you run it?

Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.

File	Quant	Size	Est. VRAM	Verdict on RTX 4090 · 24 GB
model.safetensors	bf16	6.7 GB	8.3 GB	✅ Runs comfortably

Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.

Specifications

Architecture: DeepseekOCRForCausalLM
Parameters: 3.3 B
Tensor type: BF16
Context length: 8,192
Vocabulary: 129,280
Layers / heads: 12 / 10
Licence: mit
First seen on the Hub: 2025-10-17
Training datasets: undisclosed
Added to our catalog: 2026-07-28

Compare with

Qwen3.5-9B 12.4 M Qwen2.5-VL-7B-Instruct 9.3 M Qwen2.5-VL-3B-Instruct 8.2 M Qwen3.5-4B 6.7 M Qwen3-VL-4B-Instruct 4.0 M Unlimited-OCR 2.7 M

Appears in

Best open coding models Commercial-use text models Models that fit in 12 GB