NAMAA-Space

Qari-OCR-v0.3-VL-2B-Instruct

Deploy Dedicated

README

License: apache-2.0

Model Description

QARI-OCR v0.3 is a specialized vision-language model fine-tuned for Arabic Optical Character Recognition with a focus on structural document understanding.
Built on Qwen2-VL-2B-Instruct, this model excels at preserving document layouts, HTML tags, and formatting while transcribing Arabic text.
It is described in detail in the paper QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation.

Key Features

📐 Layout-Aware Recognition: Preserves document structure with HTML/Markdown tags
🔤 Full Diacritics Support: Accurate recognition of tashkeel (Arabic diacritical marks)
📝 Multi-Font Handling: Trained on 12 diverse Arabic fonts (14px-100px)
🎯 Structure-First Design: Optimized for documents with headers, body text, and complex layouts
⚡ Efficient Training: Only 11 hours on single GPU with 10k samples
🖼️ Robust Performance: Handles low-resolution and degraded images

Model Performance

Table with columns: Metric, Score
Metric	Score
Character Error Rate (CER)	0.300
Word Error Rate (WER)	0.485
BLEU Score	0.545
Training Time	11 hours
CO₂ Emissions	1.88 kg eq.

Comparative Strengths

While QARI v0.2 achieves better raw text accuracy (CER: 0.061), QARI v0.3 excels in:

✅ HTML/Markdown structure preservation
✅ Document layout understanding
✅ Handwritten text recognition (initial capabilities)
✅ 5x faster training than v0.2

How to Use

Try Qari - Google Colab

You can load this model using the transformers and qwen_vl_utils library:

text

!pip install transformers qwen_vl_utils accelerate>=0.26.0 PEFT -U!pip install -U bitsandbytes

python

from PIL import Imagefrom transformers import Qwen2VLForConditionalGeneration, AutoProcessorimport torchimport osfrom qwen_vl_utils import process_vision_info   model_name = "NAMAA-Space/Qari-OCR-v0.3-VL-2B-Instruct"model = Qwen2VLForConditionalGeneration.from_pretrained(                model_name,                torch_dtype="auto",                device_map="auto"            )processor = AutoProcessor.from_pretrained(model_name)max_tokens = 2000 prompt = "Below is the image of one page of a document, as well as some raw textual content that was previously extracted for it. Just return the plain text representation of this document as if you were reading it naturally. Do not hallucinate."image.save("image.png") messages = [    {        "role": "user",        "content": [            {"type": "image", "image": f"file://{src}"},            {"type": "text", "text": prompt},        ],    }]text = processor.apply_chat_template(    messages, tokenize=False, add_generation_prompt=True)image_inputs, video_inputs = process_vision_info(messages)inputs = processor(    text=[text],    images=image_inputs,    videos=video_inputs,    padding=True,    return_tensors="pt",)inputs = inputs.to("cuda")generated_ids = model.generate(**inputs, max_new_tokens=max_tokens)generated_ids_trimmed = [    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]output_text = processor.batch_decode(    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]os.remove(src)print(output_text)

Try the model on Google Colab, Notebook

Training Details

Base Model: Qwen2-VL-2B-Instruct
Training Data: 10,000 synthetic Arabic documents with HTML markup
Optimization: 4-bit LoRA adapters (rank=16)
Hardware: Single NVIDIA A6000 GPU (48GB)
Framework: Unsloth + Hugging Face TRL

BibTeX:

text

@article{wasfy2025qari,  title={QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation},  author={Wasfy, Ahmed and Nacar, Omer and Elkhateb, Abdelakreem and Reda, Mahmoud and Elshehy, Omar and Ammar, Adel and Boulila, Wadii},  journal={arXiv preprint arXiv:2506.02295},  year={2025}}

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Model Details

Model Provider

NAMAA-Space

Model Tree

Base

this model

Input Modalities

TextImage

Output Modalities

Text

Supported Functionality

Dedicated Endpoints

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

Model Description

QARI-OCR v0.3 is a specialized vision-language model fine-tuned for Arabic Optical Character Recognition with a focus on structural document understanding.
Built on Qwen2-VL-2B-Instruct, this model excels at preserving document layouts, HTML tags, and formatting while transcribing Arabic text.
It is described in detail in the paper QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation.

Key Features

📐 Layout-Aware Recognition: Preserves document structure with HTML/Markdown tags
🔤 Full Diacritics Support: Accurate recognition of tashkeel (Arabic diacritical marks)
📝 Multi-Font Handling: Trained on 12 diverse Arabic fonts (14px-100px)
🎯 Structure-First Design: Optimized for documents with headers, body text, and complex layouts
⚡ Efficient Training: Only 11 hours on single GPU with 10k samples
🖼️ Robust Performance: Handles low-resolution and degraded images

Model Performance

Table with columns: Metric, Score
Metric	Score
Character Error Rate (CER)	0.300
Word Error Rate (WER)	0.485
BLEU Score	0.545
Training Time	11 hours
CO₂ Emissions	1.88 kg eq.

Comparative Strengths

While QARI v0.2 achieves better raw text accuracy (CER: 0.061), QARI v0.3 excels in:

✅ HTML/Markdown structure preservation
✅ Document layout understanding
✅ Handwritten text recognition (initial capabilities)
✅ 5x faster training than v0.2

How to Use

Try Qari - Google Colab

You can load this model using the transformers and qwen_vl_utils library:

text

!pip install transformers qwen_vl_utils accelerate>=0.26.0 PEFT -U!pip install -U bitsandbytes

python

from PIL import Imagefrom transformers import Qwen2VLForConditionalGeneration, AutoProcessorimport torchimport osfrom qwen_vl_utils import process_vision_info   model_name = "NAMAA-Space/Qari-OCR-v0.3-VL-2B-Instruct"model = Qwen2VLForConditionalGeneration.from_pretrained(                model_name,                torch_dtype="auto",                device_map="auto"            )processor = AutoProcessor.from_pretrained(model_name)max_tokens = 2000 prompt = "Below is the image of one page of a document, as well as some raw textual content that was previously extracted for it. Just return the plain text representation of this document as if you were reading it naturally. Do not hallucinate."image.save("image.png") messages = [    {        "role": "user",        "content": [            {"type": "image", "image": f"file://{src}"},            {"type": "text", "text": prompt},        ],    }]text = processor.apply_chat_template(    messages, tokenize=False, add_generation_prompt=True)image_inputs, video_inputs = process_vision_info(messages)inputs = processor(    text=[text],    images=image_inputs,    videos=video_inputs,    padding=True,    return_tensors="pt",)inputs = inputs.to("cuda")generated_ids = model.generate(**inputs, max_new_tokens=max_tokens)generated_ids_trimmed = [    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]output_text = processor.batch_decode(    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]os.remove(src)print(output_text)

from PIL import Imagefrom transformers import Qwen2VLForConditionalGeneration, AutoProcessorimport torchimport osfrom qwen_vl_utils import process_vision_info   model_name = "NAMAA-Space/Qari-OCR-v0.3-VL-2B-Instruct"model = Qwen2VLForConditionalGeneration.from_pretrained(                model_name,                torch_dtype="auto",                device_map="auto"            )processor = AutoProcessor.from_pretrained(model_name)max_tokens = 2000 prompt = "Below is the image of one page of a document, as well as some raw textual content that was previously extracted for it. Just return the plain text representation of this document as if you were reading it naturally. Do not hallucinate."image.save("image.png") messages = [    {        "role": "user",        "content": [            {"type": "image", "image": f"file://{src}"},            {"type": "text", "text": prompt},        ],    }]text = processor.apply_chat_template(    messages, tokenize=False, add_generation_prompt=True)image_inputs, video_inputs = process_vision_info(messages)inputs = processor(    text=[text],    images=image_inputs,    videos=video_inputs,    padding=True,    return_tensors="pt",)inputs = inputs.to("cuda")generated_ids = model.generate(**inputs, max_new_tokens=max_tokens)generated_ids_trimmed = [    out_ids[len(in_ids) :] for in_ids, out_ids in zip(inputs.input_ids, generated_ids)]output_text = processor.batch_decode(    generated_ids_trimmed, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]os.remove(src)print(output_text)

Try the model on Google Colab, Notebook

Training Details

Base Model: Qwen2-VL-2B-Instruct
Training Data: 10,000 synthetic Arabic documents with HTML markup
Optimization: 4-bit LoRA adapters (rank=16)
Hardware: Single NVIDIA A6000 GPU (48GB)
Framework: Unsloth + Hugging Face TRL

BibTeX:

text

@article{wasfy2025qari,  title={QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation},  author={Wasfy, Ahmed and Nacar, Omer and Elkhateb, Abdelakreem and Reda, Mahmoud and Elshehy, Omar and Ammar, Adel and Boulila, Wadii},  journal={arXiv preprint arXiv:2506.02295},  year={2025}}

@article{wasfy2025qari,  title={QARI-OCR: High-Fidelity Arabic Text Recognition through Multimodal Large Language Model Adaptation},  author={Wasfy, Ahmed and Nacar, Omer and Elkhateb, Abdelakreem and Reda, Mahmoud and Elshehy, Omar and Ammar, Adel and Boulila, Wadii},  journal={arXiv preprint arXiv:2506.02295},  year={2025}}