@karpathy Maybe you will also be interested in our concurrent work from https://t.co/6sMIQgfyIg Team 👋 Similar to DeepSeek-OCR, today we’re releasing Glyph, a visual-text compression paradigm that turns long text into images and lets a VLM read them, achieving ~3x compression rate! https://t.co/0qDorvXK8J

Maybe you will also be interested in our concurrent work from Z.ai Team 👋 Similar to DeepSeek-OCR, today we’re releasing Glyph, a visual-text compression paradigm that turns long text into images and lets a VLM read them, achieving ~3x compression rate! Paper: arxiv.org/pdf/2510.17800