Page images stay local — model weights download on opt-in (size disclosed above the gate)
Granite-Docling
Layout-aware DocTags via Transformers.js + WebGPU. Not full Python Docling.
1. Download model (~1 GB)
Loads onnx-community/granite-docling-258M-ONNX via Transformers.js on
WebGPU. Weights from Hugging Face / CDN — page images stay local.
Model not loaded
2. Image or PDF
Drop a page image or PDF here or click to browse
PNG, JPG, WEBP, or PDF — one PDF page at a time
One page at a time — batch-all would melt RAM with ~1 GB VLM.