Today, we’re launching Reve 2.0, the best 4K image model in the world.
We invented a new way to generate and edit any image using precise layouts. For the first time, it’s possible to create images you can touch.
The level of detail possible in the Reve 2.0 model is insane — now it's possible to create designs with fine print and micrographics using straight image generation
Finally testing out the new @reve model
> very realistic vibes
> perfect reference image adherence
> editing is a breeeeeze
> object manipulation is magical
> works better with shorter prompts
> text works impressively well
ps: only the guy in the Reve shirt is real, the rest is ai
"Alan Kay famously said that people who are serious about software should make their own hardware. At @reve , we believe the same principle applies to creativity: companies that are truly serious about creative tooling should train their own models" Love this!
If you ever wondered, this is what's happening inside the mind of an image diffusion model. Great tool by @reve. The new Reve 2.0 model is great and even greater is the segmentation layer on top of the generated model.
Recently joined the @reve design team, good times with @oscardumlao@Dianadotlu@_bryceking nerding out on all the details
Try out the latest model at https://t.co/sokjckM6ue
Three new models entered the Image Arena Top 10 this past month (Text-to-Image):
- #2 Reve 2.0 by @Reve (1,273), behind only GPT Image 2.
- #4 MAI-Image-2.5 by @MicrosoftAI (1,253).
- #9 Ideogram 4.0 Quality by @Ideogram_ai enters at #9 (1,204). And the only open-weights model in the top 10.
Reve 2.0 and MAI-Image-2.5 displaced their own predecessors, as both previous generations dropped out of the Top 10 with these improvements.
5/ Reve 2.0 may be the most interesting image launch this week.
The key idea: plan the image as structured layout/code first, then render native 4K. More controllable, more editable, more agent-native.
https://t.co/1Rm0OrBUDC
Extreme precision + 4K (16MP) + minimal degradation + a powerful and intuitive editor makes @reve 2.0 the best model in the world for graphic design, full stop.
All these days i thought to build a text to image model
- describe image using VLMs in 100 different ways
- collect as much data as possible for pretraining
- Do RLHF or preference optimization
- Fine tune on high quality data.
- Distill models
- Try AR or diffusion or flow
then @reve changed the way we think about all of this with layouts.
Just tried the new Reve 2.0 by @reve
the layout-based 4K image model that’s climbing the Text-to-Image arena 🔥
Pushed it hard and created a full Professional Vogue-style photoshoot.
Clean editorial lighting, high-fashion poses, flwless skin texture, and insane regional control.
The way it handles composition and detail at 4K is next level.
Storyboard-first workflow still holds strong. Using reve model speed up the process to create more storyboards
Images first → video pipeline coming next.
Drop ❤️ if you’re impressed
Comment “VOGUE” if you want the full prompt + pipeline breakdown
#Reve2.0 #AIPhotoshoot #VogueStyle
#AIFilmmaking 🇮🇳
level of control i dream of having with gen ai tools:
i should be able to say why a reference matters and what should be extracted from it.
excited for @reve
The past two days I was at @magnific's @upscaleconf (which was quite excellent, met so many people who are building creative things with AI) and EVERYONE was talking about this new model.