This is a 256M-parameter preview version of a document understanding model, specifically designed for document structure parsing and content extraction tasks, supporting the conversion of image documents into structured data.

MultimodalTransformersEnglish

SmolDocling-256M-preview-mlx-bf16 is a lightweight vision-language model converted from SmolDocling-256M-preview, specifically designed for document image understanding and conversion. This model is optimized using the MLX framework and supports converting document images into the structured DocTags format.

MultimodalTransformersEnglish

This model is converted from ds4sd/SmolDocling-256M-preview to the MLX format, supporting image-text-to-text tasks.

MultimodalTransformersEnglish

SmolDocling is an ultra-compact multimodal vision-language model specifically designed for efficient document conversion. Based on the Idefics3 architecture, this model achieves powerful document understanding capabilities with a parameter scale of 256M. It supports the extraction of various document elements such as text, tables, formulas, and code from images and is fully compatible with the Docling ecosystem.

MultimodalTransformersEnglish