Upload images in Ask Lex
under review
D
Dylan Young
I'd love to be able to generate an image from Midjourney, Flux, ChatGPT directly into Lex.
Fred Rivett
under review
Thanks for the feedback Marysol! Just so I'm sure I get you, do you mean being able to generate an image in both Ask Lex and within the editor?
If so that makes sense! It's probably not top priority for us right now, but I could see us hooking into midjourney or similar at some point.
F
Federico E
Fred Rivett I would think Marysol means "reading" an image. I have tried using Ask Lex to read images ("process" them), and I get the impression it gets a general idea of the image, but without any details.
M
Marysol Moura
Federico E Exactly! For example, Lex being able to recognize an image in my document, or in Chat.
J
Jeremy Gentles
Fred Rivett This is an important one for me as well. Being able to,
1) Upload an image for analysis. For instance, i often request the LLM estimate values from a bar graph or something similar.
2) This should also work inside a pdf (or markdown file). In most chat interfaces, I can ask the LLM to "analyze Figure 2" and it will find Figure 2 and provide feedback. This seems to work better with Gemini and Claude vs. ChatGPT.
Right now it seems that Lex may just parse and send the text from the PDF.
Fred Rivett
All good feedback thanks Jeremy, agree these would be nice to add and something I hope we’ll add at some point!