K
pending
An Open-Source Whiteboard Where AI Reads Your Handwriting in Real Time
Grounded / Real
Inflated / Uruttu
Original Content
Someone just open-sourced an AI whiteboard where you handwrite and it reponds in real time.
It's called PenEcho. You sketch equations, draw diagrams, scribble notes wherever they make sense. When you pause, it sends that region to the model. The response appears beside your work. Not in a chat window.
On the whiteboard itself.
The canvas is 20,000 x 20,000 pixels but only loads tiles where ink exists. Each request costs a few cents. Runs locally. Supports Claude, OpenAI, and Codex.
The builder said something that stuck with me: six months ago, no model could have done this. Not the handwriting recognition.
The fact that it understands unfinished equations, rough diagrams, and the spatial relationship between them. It infers what you meant from incomplete marks.
Link: https://lnkd.in/e9BUnSE2
This is how research should work. Think on a whiteboard. AI thinks with you. No copy-pasting into a chat box. No breaking your train of thought.
It's called PenEcho. You sketch equations, draw diagrams, scribble notes wherever they make sense. When you pause, it sends that region to the model. The response appears beside your work. Not in a chat window.
On the whiteboard itself.
The canvas is 20,000 x 20,000 pixels but only loads tiles where ink exists. Each request costs a few cents. Runs locally. Supports Claude, OpenAI, and Codex.
The builder said something that stuck with me: six months ago, no model could have done this. Not the handwriting recognition.
The fact that it understands unfinished equations, rough diagrams, and the spatial relationship between them. It infers what you meant from incomplete marks.
Link: https://lnkd.in/e9BUnSE2
This is how research should work. Think on a whiteboard. AI thinks with you. No copy-pasting into a chat box. No breaking your train of thought.
Validated Content
PenEcho is an open-source AI whiteboard that lets users handwrite equations, draw diagrams, and take notes anywhere on a canvas, with the AI responding directly beside the work rather than in a separate chat window. The canvas spans 20,000 by 20,000 pixels but uses sparse tile rendering so only regions containing ink are loaded, keeping the system efficient. It supports multiple model connections, including Claude, OpenAI-compatible models, and Codex, and can run locally on macOS, Windows, or via Node.js, in addition to an available cloud option. The tool is released as open source under the AGPL-3.0 license, which permits commercial use but requires that modifications served over a network also make their source code available; a separate commercial license is offered for cases that can't meet those terms. Its core capability lies in interpreting unfinished equations, rough diagrams, and the spatial relationships between handwritten elements to infer intent from incomplete input.