For a fast model, this is amazing!

Faster models are here. reve-edit-fast and reve-remix-fast are now available, and cheaper than ever.

reve.com has released reve-edit-fast and reve-remix-fast. Not only are they much faster than the previous images, but they also cost less! api.reve.com/console/cd5a515…

City policy outcomes, long term: Free buses -- improves access, reduces congestion, city gets better! City run daycare -- improves access, more available time, city gets better! Rent control -- rental supply dries up, only condos are built, city gets worse!

testing out the new #Reve model from @higgsfield_ai — handled reflections, color, and glass depth way better than I expected.

Really happy to work with the folks at FAL! They bend over backwards to be helpful. You can try our new *fast* edit models through their API right now!

The PVKK developers saw Steel Battalion and said "hold my beer!" piped.video/shorts/yKmpQ_m1M…

“What is reality?” - Every Human Ever Inspired by @reve’s mantra of reimagining reality, we decided to give answering this question a try in our newest short film. (Visuals generated with Reve, obviously)

What's your halloween costume? lab.reve.com/halloween/share…

Make your images in Reve, and turn them into video right there! Most controllable image tool -> most controllable video.

3/ Image Reference Combine prompts with image references for highly specific results. You can create your video by generating a new reference image, uploading an image, or choosing from previously attached images, then simply describe what you want to see. Reve brings the elements together seamlessly.

I'm trying to figure out how they get 20 kilograms of payload and 70 kilograms of max weight lift into a battery powered humanoid robot weighing 30 kilograms. "Tendon activated" isn't enough. Is there a tech primer somewhere?

this post is complete misinformation LLMs are lossy compressors! of *training data*. LLMs losslessly compress *prompts*, internally. that’s what this paper shows. source: i am the author of “Language Model Inversion”, the original paper on this

OpenAI research is so AGI pilled we bet our whole codebase that we’ll hit superhuman coding before tech debt bankruptcy

I glanced through some of your patents and honestly I don't think they should have been awarded. I will explain my reasoning in this thread.

A fantastic day of talks exploring the future of design & AI at the Interpolations Conference!

people are going to have to come to terms that the gentleman’s agreement with google on scrape for referral was not actually legally binding and has been allowed for 30 years by publishers. how do you fairly change that?

New breakthrough quantum algorithm published in @Nature today: Our Willow chip has achieved the first-ever verifiable quantum advantage. Willow ran the algorithm - which we’ve named Quantum Echoes - 13,000x faster than the best classical algorithm on one of the world's fastest supercomputers. This new algorithm can explain interactions between atoms in a molecule using nuclear magnetic resonance, paving a path towards potential future uses in drug discovery and materials science. And the result is verifiable, meaning its outcome can be repeated by other quantum computers or confirmed by experiments. This breakthrough is a significant step toward the first real-world application of quantum computing, and we're excited to see where it leads.

I feel seen!

I am sure they would all agree to donate it to charity and it would be the most insane game of all time.