Instructions to use OdiaGenAIOCR/odia-ocr-qwen-finetuned_v3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use OdiaGenAIOCR/odia-ocr-qwen-finetuned_v3 with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Request to List odia-ocr-qwen-finetuned_v3 as a Community Model
Hello Hugging Face Team,
We would like to request adding our model, odia-ocr-qwen-finetuned_v3, to the Community Models list.
Model Details:
Repository: OdiaGenAIOCR / odia-ocr-qwen-finetuned_v3
Task: Odia OCR using Vision–Language Model (VLM)
Architecture: Qwen2.5-VL-7B, fine-tuned with LoRA (PEFT)
Dataset: 58K+ Odia image-text pairs (synthetic + real OCR datasets)
Capabilities: Paragraph-level Odia text extraction across fonts and layouts
This model is intended to support research, document digitization, and AI development for Odia and Indic languages.
We would be grateful if the Hugging Face team could review this request and include the model in the Community Models list.
Thank you!
Team OdiaGenAI OCR Team