SmolDocling is a multimodal Image-Text-to-Text model designed for efficient document conversion. It retains Docling's most popular features while ensuring full compatibility with Docling through seamless support for DoclingDocuments.
Features
27x smaller than comparable models
SmolDocling
SmolDocling beats standard document processing and bigger models in several ways.
Even with a mere 256 million parameters, SmolDocling performs like models that are 27 times bigger.
Process a page in only 0.35 seconds using an NVIDIA A100 GPU and small amounts of resources.
The DocTags format provides a clear, machine-friendly format that keeps the document's structure.
More precise and dependable than larger models, with fewer incidents of fabricated information.
Applications
SmolDocling performs well in various document processing situations and sectors.
Handle invoices, receipts, financial reports, and agreements with great precision, keeping the original formatting.
Pull out data from legal agreements, court files, and case documents with high accuracy.
Process medical documents, lab findings, and patient records while retaining formatting and layout.
Get data from research papers, including complex tables, formulas, and references for review.
Innovative Technology
SmolDocling is a leap forward in document processing, skillfully merging efficiency with superior performance.
<doc> <block type="heading" level="1"> <loc x="50" y="100" w="500" h="60"> SmolDocling Documentation </loc> </block> <block type="paragraph"> <loc x="50" y="180" w="500" h="100"> SmolDocling is an efficient document processing model with... </loc> </block> </doc>
User Reviews
Find out how SmolDocling is revolutionizing document handling processes for companies globally.
Financial Tech Solutions
Academic Research Institute
Legal Analytics Partners
FAQS