Introducing Images 2.6

Re:Introducing Images 2.6 by jjcm on Wednesday November 04, 2026 @07:33PM (#50373420)

I've been playing with it for the last hour or so. I immediately updated my platform to use both since 2.5 is my primary model (i run a prompt->design site).

After playing with it for an hour, I'm rolling back 2.6-flare to 2.5-flare. I've found 2.6-flare specifically to be too simplistic in its approaches.

Examples: https://diffui.ai/app/canvas/178db5d5-ff87-4b43-aa98-2897773...

2.6-sunburst seems solid however.

[ Parent | on HN ]

Re:Introducing Images 2.6 by Computer0 on Wednesday November 04, 2026 @07:46PM (#50373536)

I preferred 2.6 in all examples supplied!

[ Parent | on HN ]

Re:Introducing Images 2.6 by jjcm on Wednesday November 04, 2026 @07:48PM (#50373559)

Can you explain more? In two of the examples the images from 2.6 are inherently broken, ie: https://image.non.io/161d8f02-fc2a-40e0-91b2-9d27ceb8fef1.we...

Why do you prefer the outputs with errors over the ones without?

[ Parent | on HN ]

Re:Introducing Images 2.6 by jjcm on Wednesday November 04, 2026 @08:26PM (#50373861)

Update: based on info from design arena, it looks like gpt-image-2.6 was post-trained to output less text: https://x.com/DesignArena/status/2118124381155303918

This may be why text elements in particular appear more "zoomed in", and why it's providing less detail for UI design.

[ Parent | on HN ]

Re:Introducing Images 2.6 by ChickeNES on Wednesday November 04, 2026 @08:55PM (#50374107)

Oh thank god, the awful inspirational text shoved needlessly into everything has been excised.

[ Parent | on HN ]

Re:Introducing Images 2.6 by DavCreator on Thursday November 05, 2026 @06:51AM (#50378059)

https://xnotme.com/DesignArena/status/2118124381155303918

[ Parent | on HN ]