The First Open Weight Model for Deterministic Work: Interfaze 1 Lite - Interfaze
interfaze-1-lite is the first open-weight model built for deterministic work like OCR, speech-to-text and classification, with typed output, boxes and...
Python Library
Summary
CommonForms is an open-source Python package with ML models (FFDNet-S/L) that automatically detects form fields in PDFs and converts them into fillable forms, available via CLI and API.
Highlights
- Pip-installable package with both CLI and Python API for converting PDFs to fillable forms
- Two model sizes (FFDNet-S and FFDNet-L) trained on a dedicated form field detection dataset
- Supports GPU inference, confidence thresholds, and options to keep existing fields or use signature fields
- Backed by a research paper and a large, diverse dataset hosted on HuggingFace
- Hosted models available at detect.semanticdocs.org for direct use
auto-generated
Context
Audience
Developers and document processing engineers who need to automate PDF form creation or fillable form detection
DomainComputer Vision / Document Processing
Formatopen-source software repository
Accessopen source, free to install via pip
Source Overview
CommonForms — open models to auto-detect PDF form fields - jbarrow/commonforms
Topics
Related
FFDNet-S and FFDNet-L modelsCommonForms dataset on HuggingFacearXiv:2509.16506detect.semanticdocs.orgPDF form automation
