numind/NuExtract
NuExtract by NuMind is a 4 billion parameter language model, fine-tuned from Microsoft's Phi-3-mini-4k-instruct, specifically designed for structured information extraction. It excels at extracting data from text into a specified JSON format, ensuring all extracted text is present in the original input. This model is purely extractive and supports a 4096 token context length, making it suitable for precise data retrieval tasks.