Grok Imagine 1.5 Preview is now available on Grok APIs and has also landed in the top 1 spot on Arena. Have you tested it already?
CREATIVE AI
-
Only unreleased model on LM Arena Preview follows minimalistic instruction
By
–
Technically, "unreleased" model is already available on LM Arena in Preview. And it was the only one that followed "minimalistic" instruction properly. – Image 1 – MAI Image 2.5 – Image 2 – GPT Image 2
– Image 3 – Nano Banana 2
– Image 4 – MAI Image 2 -
Sensitivity to generative model patterns and craving raw expression
By
–
I have become super sensitive to the patterns introduced by many of the current generative models. I get triggered by contrastive negation and sloppy clip art. I want to see the output of your mind, raw existential expression, not filler
-
Amazon plans to launch AI content marketplace
By
–
Amazon plans to launch AI content marketplace
#AI #AIio #AIInnovation #ML #DataScience #Futureofwork @timnitgebru @oriolvinyalsml @ceobillionaire @soumithchintala @waitin4agi_ @sallyeaves @bernardmarr -
Image comparison: unreleased model, Nano Banana Pro, GPT Image 2
By
–
Which image do you like the most? One of them is from the unreleased model, one is Nano Banana Pro and one is GPT Image 2. Prompt below. Model reveal later today.
-
Renaming AI Artist list to Creative with AI
By
–
I renamed my "AI Artist" list to "Creative with AI" because I don't really care about your arguments that AI can't create art: https://
x.com/i/lists/169702
3939338519013?s=20
… AI is for people who are creating things. -
Renamed AI Artist list to Creative with AI, acknowledges movement’s pace
By
–
I renamed my AI Artist list to "Creative with AI" because I don't care about the argument anymore. You are right that the movement won't slow down:
-
Vibe coding empowers non-programmers to create with AI
By
–
Vibe coding isn’t really about writing code — it’s about changing who has the power to create AI-powered development tools are making it possible for people without traditional programming backgrounds to build applications, automate workflows and bring ideas to life faster
-
PrismML’s Bonsai Image 4B: Powerful image generation model now on your iPhone
By
–
PrismML just shrunk a 4B image model down to fit on your iPhone.
— AlphaSignal AI (@AlphaSignalAI) 30 mai 2026
Most image generation today runs in the cloud. Every prompt becomes a remote request with latency and cost.
PrismML released Bonsai Image 4B to change that.
It runs diffusion inference directly on phones and… pic.twitter.com/J09ZSoqAHMPrismML just shrunk a 4B image model down to fit on your iPhone. Most image generation today runs in the cloud. Every prompt becomes a remote request with latency and cost. PrismML released Bonsai Image 4B to change that. It runs diffusion inference directly on phones and
-
Two-step prompting: camera move first, then add character
By
–
It’s still pretty good! Maybe the simple solution is to first prompt it to get the camera move right, without the character – just the drone point of view. Then take that video and prompt it to add the character with text or img ref.