Instructions to use Gramscii-IT/DeepSeek-OCR-2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Gramscii-IT/DeepSeek-OCR-2-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
# Run inference directly in the terminal:
llama cli -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Use Docker

docker model run hf.co/Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

LM Studio
Jan

vLLM

How to use Gramscii-IT/DeepSeek-OCR-2-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Gramscii-IT/DeepSeek-OCR-2-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gramscii-IT/DeepSeek-OCR-2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Ollama
How to use Gramscii-IT/DeepSeek-OCR-2-GGUF with Ollama:
```
ollama run hf.co/Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
```
Unsloth Desktop
Docker Model Runner
How to use Gramscii-IT/DeepSeek-OCR-2-GGUF with Docker Model Runner:
```
docker model run hf.co/Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0
```

Lemonade

How to use Gramscii-IT/DeepSeek-OCR-2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Gramscii-IT/DeepSeek-OCR-2-GGUF:Q8_0

Run and chat with the model

lemonade run user.DeepSeek-OCR-2-GGUF-Q8_0

List all available models

lemonade list

Atomic Chat

DeepSeek-OCR-2 for SDG

DeepSeek-OCR-2 Q8_0 language weights and BF16 vision projector, distributed for Gramscii SDG local document reading. Both artifacts are required. Inference runs locally; documents are not uploaded to Hugging Face.

Original model: deepseek-ai/DeepSeek-OCR-2, Apache 2.0. GGUF conversion: sabafallah/DeepSeek-OCR-2-GGUF. These files are byte-identical to that pinned conversion; Gramscii does not claim authorship of the model or conversion.

Use llama.cpp with DeepSeek-OCR-2 support and the deepseek-ocr chat template, Jinja disabled, flash attention disabled. SDG owns download verification, process lifecycle and task-specific prompts. Its model manifest pins this repository's immutable revision and artifact checksums.

Artifact	Bytes	SHA-256
deepseek-ocr-2-q8_0.gguf	3126139808	`4d82d12b949a6256a66eed9874a8177a9a25d1fe1b0650dfaaf34c9b92d9b911`
mmproj-deepseek-ocr-2-bf16.gguf	929037632	`b65e8460acc82dd4e8546206c5abd1abed9c5c582223731ce1585689d95f5cdb`

OCR text and coordinates require review. This model does not guarantee pixel-identical PDF reconstruction or correct identification of editable form fields.

Downloads last month: 276

GGUF

Model size

3B params

Architecture

deepseek2-ocr

Hardware compatibility

8-bit

Model tree for Gramscii-IT/DeepSeek-OCR-2-GGUF

Base model

deepseek-ai/DeepSeek-OCR-2

Quantized

(10)

this model