Skip to content
Better HN
Top
New
Best
Ask
Show
Jobs
Search
⌘K
undefined | Better HN
0 points
ryanackley
1y ago
0 comments
Share
The technology that allows an LLM to "see" images and video is completely different though. It's not what is being trained on common crawl.
0 comments
default
newest
oldest
brrrrrm
1y ago
not really. embeddings are embeddings. check out llava
j
/
k
navigate · click thread line to collapse