ChatGPT Images - Powered by GPT Image 1.5: Faster, smarter, precise
ChatGPT Images is powered by the new GPT Image 1.5 model. It delivers precise edits that preserve lighting and facial likeness, stronger instruction following, and 4x faster generation. Available now in ChatGPT and via API.
Replies
Wow, OpenAIs new image editing is incredible! The precise lighting preservation is a game-changer. How does the API handle complex scene edits with multiple light sources?
We have already tested it and appreciate how ChatGPT is increasingly approaching Canvas in terms of capabilities. We now use the ChatGPT collection exclusively to publish social media graphics.
just tried it today, I like that it has fixed the consistency issue (in many cases).
but still not as good as my main image model -- nano banana.
for example, this model will do extra editing which is not part of my instructions, and the worst thing is that it can't revert things back ...
Ah yes, the classic “I asked for a red button, and it redesigned my whole UI in Comic Sans” problem.
Wow, this is incredible! The GPT Image 1.5 models faster generation is a game changer. How does it handle complex scene understanding for edits?
Anyone else noticing improvements in text rendering?