numind/NuExtract-1.5-tiny

NuExtract-1.5-tiny by NuMind is a 0.5 billion parameter language model, fine-tuned from Qwen2.5-0.5B, specifically designed for structured information extraction from long documents. It excels at extracting data into a JSON template across multiple languages including English, French, Spanish, German, Portuguese, and Italian. The model prioritizes pure extraction, ensuring generated text is present in the original source, and supports a 32768 token context length.