PatentFig AI is a browser-based platform that uses artificial intelligence to generate patent-oriented figures from text descriptions, sketches, product photographs, and 3D model references.

Key features and functions include:

Multi-Format Figure Generation
The platform produces line art, block diagrams, flowcharts, schematics, exploded views, cross-sections, and multi-view sets for both utility and design patent applications.

Natural-Language Editing
Users can revise generated figures through plain-language instructions (for example, adjusting line weight or adding a sectional view) rather than manual redrawing.

Reference Numeral Consistency
The system maintains consistent reference numerals and labeling across related figures within a project.

Version Control
Each generation is tracked as a version (V1, V2, etc.), with sub-versions recorded for subsequent edits, allowing users to compare or revert to earlier drafts.

Figure Compliance Checking
A built-in checking tool reviews figures against formatting conventions associated with the USPTO, EPO, CNIPA, JPO, KIPO, and PCT/WIPO systems, evaluating factors such as DPI, margins, and reference numeral usage.

Multi-Office Formatting
Users can switch a single set of source figures between office-specific formatting workflows for the USPTO, EPO, CNIPA, JPO, KIPO, and PCT/WIPO.

Export Options
Figures can be exported at 300 or 600 DPI and converted to SVG, DXF, or vector PDF, in addition to PDF, PNG, and TIFF formats.

Project Organization
Figures are organized by patent application, with each project tracking associated figure sets, versions, and compliance status.