When you purchase through links on our site, we may earn an affiliate commission.Heres how it works.
DeepSeekis on a roll.
If you’re curious to try it for yourself, you canaccess the model at HuggingFace here.
Images generated using the Janus Pro artificial intelligence model(Image credit: Janus Pro / Tom’s Guide)
This is not a typical combination with conventional models.
They call it unified multimodal.
Think aboutwhat Stable Diffusion was like in 2023and you’ll know what I’m talking about.
Images generated using the Janus Pro artificial intelligence model(Image credit: Janus Pro / Tom’s Guide)
Its a shame, but I guess innovation often comes with a price.
you’re able to see the examples below.
The good news is the image vision seems to work fine.
(Image credit: Janus Pro / Tom’s Guide)
Bottom line
So where does that leave us?
Combining image generation with the ability to read images is a nice feature.
However, the report card on this attempt must read “could try harder.”
(Image credit: Janus Pro / Tom’s Guide)
Its a total mystery.
I suggest they maybe take a trip back to the drawing board?
More from Tom’s Guide
(Image credit: Janus Pro / Tom’s Guide)
