DeepSeek OCR 2 needs about 2.7 GB of memory at Q4_K_M (a 2.0 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 3 quants, from 2.0 GB (Q4_K_M) to 5.9 GB (BF16).
DeepSeek OCR 2 – 2.93B parameter OCR model
DeepSeek OCR 2 is a 2.93B parameter OCR model built on the deepseek2-ocr architecture. It supports dynamic resolution and visual causal flow. The model is available in GGUF format for llama.cpp PR #20975.
Good at
- OCR on variable resolution images
- Dynamic visual token handling
- Efficient inference with GGUF
Limits
- No vision support on standard llama.cpp main branch
- License not specified
- No explicit context token limit
Best for
- Document image OCR
- PDF text extraction
- Batch image text recognition
Who it’s for. Developers needing OCR capabilities in local LLM deployments. Suitable for projects that require dynamic resolution handling and visual causal flow.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 2.0 GB | 2.7 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 → |
For 8 GB VRAM, use Q4_K_M (1.95 GB file, ~2.7 GB memory at 8k context). For 12 GB VRAM, Q8_0 (3.13 GB file, ~4 GB memory). For larger memory budgets, BF16 (5.88 GB file, ~7.1 GB memory).
What GPU do you need for DeepSeek OCR 2?
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 2 quant should you choose?
Q4_K_M is the usual starting point for DeepSeek OCR 2: it balances file size and quality well. Here it is a 2.0 GB file.
The Q8_0 file is 3.1 GB, about 1.6 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about DeepSeek OCR 2
How much VRAM does DeepSeek OCR 2 need?
About 2.7 GB at Q4_K_M with an 8K context window. That is the 2.0 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 2 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 2 GGUF?
It depends on the quant: Q4_K_M is 2.0 GB, Q8_0 is 3.1 GB, BF16 is 5.9 GB.
Where can I download DeepSeek OCR 2 GGUF?
The files are on Hugging Face in the sabafallah/DeepSeek-OCR-2-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 2 on 8 GB VRAM?
Yes, using the Q4_K_M quantized model which requires about 2.7 GB of VRAM at 8k context.
Does this model support dynamic resolution?
Yes, it supports dynamic resolution from 0–6×768×768 up to 1×1024×1024 with visual tokens.
Is the model compatible with the main llama.cpp branch?
No, it requires the llama.cpp PR #20975 branch to run the GGUF files.