Get the latest tech news

Karpathy on DeepSeek-OCR paper: Are pixels better inputs to LLMs than text?


We’ve detected that JavaScript is disabled in this browser. Please enable JavaScript or switch to a supported browser to continue using x.com.

None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of Text

Text

Photo of LLMs

LLMs

Photo of karpathy

karpathy

Related news:

News photo

Drawing Text Isn't Simple: Benchmarking Console vs. Graphical Rendering

News photo

LLMs can get "brain rot"

News photo

Neural audio codecs: how to get audio into LLMs