Get Started
Get started with the Interactive Ink Canvas — draw, edit and recognize text, math and shapes.
Explore handwriting recognition demos — Interactive Ink Canvas, Ink Canvas (v2), and advanced integrations.
Get started with the Interactive Ink Canvas — draw, edit and recognize text, math and shapes.
Interactive Ink Canvas feature tour — inspect symbols, history, selection and backend rendering.
All 6 gesture types with visual reference and gesture menu configuration.
Configure canvas defaults, behavior overrides, and initial settings.
Customize background, menu panels, and input appearance via cssVars CSS custom properties.
Show or hide specific menus and toolbar actions without custom code.
Replace built-in menus with fully custom menu classes.
Replay a recorded set of strokes on the canvas with its original timing, with play/pause/speed control.
Programmatically create and inject ink symbols into the canvas.
Programmatic zoom and pan — set zoom level, fit to content, and shift the viewport via the API.
Programmatic symbol selection — select all, select by ID, react to selection events, zoom to selection.
Bird's-eye overview panel for navigating large ink canvases.
Keyboard navigation and shortcut bindings for hands-free editing.
Real-time canvas state snapshot and timestamped action log — symbols, selection, undo/redo, zoom, and clipboard.
Live connection state indicator with automatic reconnect handling.
Handwriting recognition embedded inside a TLDraw collaborative canvas (React).
Switch between numeric evaluation and symbolic computation modes.
Track variables across linked math blocks for dependency-aware computation.
Surround a math expression to compute, evaluate functions, set variables, or check the solver diagnostic.
Declare variables as ink expressions, substitute them across blocks, and manage global fallbacks via the variable canvas.
Math equation recognition with the new REST v2 API — no WebSocket needed.
Shape detection and vectorization — circles, arrows, polygons, and more.
Raw digital ink recognition — no explicit content type segmentation.
Text recognition using the REST v2 API — simple batch handwriting to text.
Vertical Japanese text recognition — tategumi layout with REST v2.
One independent InkCanvas instance per question field — validate each answer via export and tally a score.
A grid of independent InkCanvas instances — one participant per row, one grade per criterion column, plus a text comment.