←back to thread

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

(twitter.com)
237 points JnBrymn | 1 comments | 21 Oct 25 17:43 UTC | HN request time: 0.247s | source
Show context
dgfitz[dead post] ◴[23 Oct 25 01:23 UTC] No.45677161[source]▶
[flagged]
1. scotty79 ◴[23 Oct 25 02:02 UTC] No.45677380[source]▶
It's kind of beautiful that they can actually do that.