Christian Cantrell

Six months ago, we launched a preview of our first model. Today, we're launching the new Reve. But it's really four products in one: 1. An image creator and remixer. 2. A drag-and-drop image editor. 3. Your own creative assistant. 4. For the first time, an API. Let's start with remix...

make everything cinematic in @reve

One shotting a cinematic photo edit with @reve annotations

@sampullara I think that was the idea behind the yoke and no dash display in the first place. When I drive my Tesla, I almost never drive.

now that i don’t use the steering wheel very much maybe the yoke would be fine

Get picture perfect with annotations

Write fewer prompts. Visual work should feel visual. Just draw what you want to see.

Annotations in Reve. Show, don't tell.

Annotations are one of the most powerful, intuitive, and delightful features we've shipped in Reve so far.

Capture the details in 4K

Reve team is killing it with the new Reve 1.5 model! Prompted beautiful 4K editorial-ready visuals below Also the @reve layers makes post production super easy to edit any element to the minuet-detail

Reve v1.5 excels at product, branding and commercial design

@_javierivero @reve Just added these things to the backlog!

That's great to hear, it's honestly vey frustrating to lose all of that grain, texture & sharpness when upscaling. Couple of other features in my wish list: - When asking to merge a style from one picture to another I have to be hyper specific to not modify any shapes/form from my original image, often taking even 2 sentences long to set this instruction and sometimes the model just creates a completely new version of the original image. This is why I still use 🍌 for this worflow. - When I select "Previously attached" from the "Attach" button it forgets my setting of "latest first" i never want to see my previously attached images ordered alphabetically - After a couple of iterations of the same frame, some details start to get messed up (hands, shadows, borders, etc) I have to open a new chat/session every time i make an interation to avoid this from happening Hope this helps!

Introducing the brand new, 4K (16MP), @reve 1.5 model. Third best model in the world, and it's still cooking. More to come...

@_javierivero @reve We're using a third-party upscaler. Rather than train our own, we're just going to generate natively at very high resolutions, starting with 4K.

@cantrell @reve Would love for Reve to don’t smooth details out when upscaling an image. It ruins the realistic feel and gives it an unmistakable AI look & feel

I'm beyond excited about this native 4k image generation model that we've worked on for months! It feels great that the image can be directly used as a wallpaper.

Reve's new text-to-image model is here. Really proud of the team to rank at #3 with the big labs. To my knowledge, we are the first lab to use native pixel space diffusion without latent autoencoder at 4k (16MP) resolution for production level image generation.

Reve v1.5 is here. Our latest image model, now with 4K resolution.

Turning a house into a home 🏠

Using emojis may actually be the most underrated prompting tool 🙅‍♂️

Reve has been redesigned. Edit anything.

Dinner for two?