Instructions to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with PEFT:

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("unsloth/qwen2-vl-2b-instruct-unsloth-bnb-4bit")
model = PeftModel.from_pretrained(base_model, "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct")

Transformers

How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct",
		"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/NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct

SGLang

How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct",
		"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 images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct",
		"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"
						}
					}
				]
			}
		]
	}'

Unsloth Studio

How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct",
    max_seq_length=2048,
)

Docker Model Runner
How to use NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct with Docker Model Runner:
```
docker model run hf.co/NAMAA-Space/Qari-OCR-0.1-VL-2B-Instruct
```

nielsr HF Staff commited on Jun 5, 2025

Commit

79f2d40

verified ·

1 Parent(s): 49d876c

Add link to paper and Github repository, update library_name

Browse files

This PR adds a link to the paper [QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation](https://huggingface.co/papers/2506.02295) and a link to the Github repository. The `library_name` is also updated from `peft` to `transformers`.

Files changed (1) hide show

README.md +11 -8

README.md CHANGED Viewed

@@ -1,21 +1,21 @@
 ---
 base_model:
 - unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit
-tags:
-- transformers
-- unsloth
-- qwen2_vl
-- trl
-- ocr
-license: apache-2.0
 language:
 - ar
 metrics:
 - bleu
 - wer
 - cer
 pipeline_tag: image-text-to-text
-library_name: peft
 ---
 # Qari-OCR-0.1-VL-2B-Instruct Model
@@ -24,6 +24,7 @@ library_name: peft
 This model is a fine-tuned version of [unsloth/Qwen2-VL-2B-Instruct](https://huggingface.co/unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit) on an Arabic OCR dataset. It is optimized to perform Arabic Optical Character Recognition (OCR) for full-page text.
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/628f7a71dd993507cfcbe587/HuUcfziXcDT_2kwDoz5qH.png)
@@ -92,7 +93,9 @@ While the Arabic OCR model demonstrates strong performance under specific condit
 These limitations should be considered when deploying the model in real-world applications to ensure optimal performance.
 ## How to Use

 ---
 base_model:
 - unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit
 language:
 - ar
+library_name: transformers
+license: apache-2.0
 metrics:
 - bleu
 - wer
 - cer
 pipeline_tag: image-text-to-text
+tags:
+- transformers
+- unsloth
+- qwen2_vl
+- trl
+- ocr
 ---
 # Qari-OCR-0.1-VL-2B-Instruct Model
 This model is a fine-tuned version of [unsloth/Qwen2-VL-2B-Instruct](https://huggingface.co/unsloth/Qwen2-VL-2B-Instruct-unsloth-bnb-4bit) on an Arabic OCR dataset. It is optimized to perform Arabic Optical Character Recognition (OCR) for full-page text.
+The model is described in the paper [QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation](https://huggingface.co/papers/2506.02295).
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/628f7a71dd993507cfcbe587/HuUcfziXcDT_2kwDoz5qH.png)
 These limitations should be considered when deploying the model in real-world applications to ensure optimal performance.
+## Code
+https://github.com/NAMAA-ORG/QARI
 ## How to Use