Page images stay local — model weights download on opt-in (size disclosed above the gate)
Florence-2
Multi-task VLM with an <OCR> task token. Not DocTags — for layout/structure prefer Granite-Docling.
WebGPU required. This browser does not expose navigator.gpu — Florence-2 cannot load here.
1. Download model (~320 MB)
Loads onnx-community/Florence-2-base-ft with dtype fp16 on WebGPU.
Runs the <OCR> task. Large download — opt-in only.
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.
3. OCR text
Need DocTags / HTML / MD? Try Granite-Docling.