←back to thread

Gemini 2.5 Flash Image

(developers.googleblog.com)
1092 points meetpateltech | 5 comments | 26 Aug 25 14:01 UTC | HN request time: 0s | source
1. asadm ◴[26 Aug 25 17:26 UTC] No.45029632[source]▶
this is amazing. I just wish models would have more non-textual controls. I don't want to TYPE my instructions. We need a better UI for editing images with AI.
replies(2): >>45030217 #>>45034901 #
2. mh- ◴[26 Aug 25 18:14 UTC] No.45030217[source]▶
Can you expand on that? What would ideal look like to you?
3. ojr ◴[27 Aug 25 02:43 UTC] No.45034901[source]▶
You want to use a mouse and keyboard and learn 20 buttons like it's 1990?
replies(2): >>45035632 #>>45055990 #
4. yborg ◴[27 Aug 25 05:00 UTC] No.45035632[source]▶
I don't want to converse with a 4 year old with the world's photo libraries at its disposal. I spent 10 minutes trying to convince the model to add a watch to a person's left arm instead of the right arm and it would not do it, it apparently could not get the idea on this particular image. If I had a drawing tool I could circle where I wanted it and say 'THERE, stupid'. Next year when we have AGI all of this will be moot of course, but for now Photoshop isn't going away.
5. asadm ◴[28 Aug 25 19:23 UTC] No.45055990[source]▶
yes! for extremely precise edits.