DeepSeek-OCR needs about 3.1 GB of memory at Q5_K (a 2.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 8 quants, from 1.7 GB (Q4_0) to 5.9 GB (F16).
DeepSeek-OCR: Lightweight OCR model for document and scene text
DeepSeek-OCR is a 2.93 billion‑parameter vision‑language model optimized for optical character recognition. It extracts text from images, preserves layout, and supports multilingual output. The model is available in GGUF format for efficient local inference.
Good at
- Multilingual OCR
- Document layout preservation
- Scene text recognition
Limits
- No vision support in the base model card
- License not specified
- No explicit context token limit
Best for
- Digitizing scanned documents
- Parsing invoices and receipts
- OCR‑based search systems
Who it’s for. Developers needing a compact OCR solution for edge devices or local deployment. The model works with NVIDIA GPUs and can be run via NexaSDK or llama.cpp.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_0 | 1.7 GB | 2.3 GB | Fits | Get file → | |
| Q5_0 | 2.0 GB | 2.8 GB | Fits | Get file → | |
| Q4_K | 2.1 GB | 2.8 GB | Fits | Get file → | |
| Q5_K | 2.3 GB | 3.1 GB | Fits | Get file → | |
| Q6_K | 2.6 GB | 3.4 GB | Fits | Get file → | |
| Q8_0 | 3.1 GB | 4.0 GB | Fits | Get file → | |
| BF16 | 5.9 GB | 7.1 GB | Fits | Get file → | |
| F16 | 5.9 GB | 7.1 GB | Fits | Get file → |
For 8 GB VRAM, Q4_0 (1.65 GB, 2.3 GB memory) or Q5_0 (2.02 GB, 2.8 GB memory) are suitable. If 12 GB VRAM is available, Q6_K (2.61 GB, 3.4 GB memory) or Q8_0 (3.12 GB, 4 GB memory) can be used. Higher precision quants require more memory.
What GPU do you need for DeepSeek-OCR?
These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is Q8_0 (3.1 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (3.1 GB file, about 4.0 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (5.9 GB file, about 7.1 GB in use).
Which DeepSeek-OCR quant should you choose?
Q5_K (2.3 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.
If memory is tight, the smallest file is Q4_0 at 1.7 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about DeepSeek-OCR
How much VRAM does DeepSeek-OCR need?
About 3.1 GB at Q5_K with an 8K context window. That is the 2.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run DeepSeek-OCR on an 8 GB GPU?
Yes, with BF16: the 5.9 GB file needs about 7.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of DeepSeek-OCR GGUF?
It depends on the quant: Q4_0 is 1.7 GB, Q5_0 is 2.0 GB, Q4_K is 2.1 GB, Q5_K is 2.3 GB, Q6_K is 2.6 GB, Q8_0 is 3.1 GB and 2 more.
Where can I download DeepSeek-OCR GGUF?
The files are on Hugging Face in the NexaAI/DeepSeek-OCR-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run DeepSeek-OCR on 8 GB VRAM?
Yes, the Q4_0 quant file is 1.65 GB and requires about 2.3 GB memory at 8k context, fitting within 8 GB VRAM.
Does the model support image-to-text conversion?
Yes, it outputs extracted text and can provide structured JSON with bounding boxes.
What license applies to the model?
The model is released under the Apache 2.0 License, allowing commercial use.