dhanushmekaka/qwen25-1.5b-invoice-extraction
The dhanushmekaka/qwen25-1.5b-invoice-extraction model is a 1.5 billion parameter Qwen2-based language model developed by dhanushmekaka. It is specifically fine-tuned for invoice extraction tasks, leveraging the Qwen2.5-1.5b-instruct-unsloth-bnb-4bit base model. This model was trained using Unsloth and Huggingface's TRL library, optimizing for faster training. Its primary strength lies in accurately extracting information from invoices.
Loading preview...
Model Overview
This model, dhanushmekaka/qwen25-1.5b-invoice-extraction, is a specialized 1.5 billion parameter language model built upon the Qwen2 architecture. Developed by dhanushmekaka, it is specifically fine-tuned for the task of invoice extraction.
Key Characteristics
- Base Model: Fine-tuned from
unsloth/qwen2.5-1.5b-instruct-unsloth-bnb-4bit. - Parameter Count: Features 1.5 billion parameters, offering a balance between performance and computational efficiency.
- Training Optimization: Training was accelerated by a factor of 2x using the Unsloth library in conjunction with Huggingface's TRL library.
- Context Length: Supports a substantial context length of 32768 tokens, allowing for processing of longer documents.
Primary Use Case
This model is explicitly designed and optimized for invoice extraction. Developers should consider this model for applications requiring automated and accurate data retrieval from invoice documents. Its fine-tuned nature suggests superior performance for this specific domain compared to general-purpose language models of similar size.