PixVerse R2
A real-time world model you can explore and change
Start new thread

PixVerse R2 - A real-time world model you can explore and change

by
•
PixVerse R2 is a real-time world model that generates continuously evolving audiovisual worlds instead of fixed video clips. It accepts text, images, audio and actions while generating, remembers what happened earlier in the session, and carries those changes forward in real time. R2 scales to longer, more coherent and controllable experiences — powering everything from interactive stories and characters to playable generative worlds.

Add a comment

Replies

Best
Maker
📌

Hey Product Hunt! 👋

We’re the PixVerse team. Meet PixVerse R2: a real-time world model you can explore and change while it’s being generated.

Explore. Change. Keep going.

  • Explore with WASD. Move through the world and choose where to go next.

  • Change it with a prompt. Introduce a new element or change the scene as it unfolds.

  • Keep going. Continue interacting in the same evolving world after each change.

A collection of R2-powered worlds is now open to explore.

How R2 works

R2 builds its world-model capabilities through continuous pretraining with Omni Causal AR, then distills that same foundation for real-time interaction. The aim is to make worlds more persistent, controllable, and scalable.

Whether you’re experimenting with interactive stories, game prototypes, or AI worlds, we’d love your feedback. What would you explore or change first?

The gallery combines AI concept scenes with an overview of R2’s architecture.

Hunter

 Shoutout to this new launch, Jaden! Keep up the great work!

Maker
Thank you Justin🫶🏻

Checked the gallery, and there are some really interesting worlds in there.
Definitely worth exploring.

Thanks and TEam....

Maker
Thanks so much, Deepak! Really glad you checked out the gallery 🙌 We’re excited to keep pushing R2 and see what kinds of worlds people build and explore with it. Appreciate the support!
congrats on the launch! pixverse has always been my go-to tool for aigc content production. Glad to see the new model and can’t wait to try it out!
Maker

 Thanks Alex！Congrats for your new career update!

​Can we export session memory state as a preset to resume generating complex scenes later?

Maker
Ohhh it’s a pretty cool perspective

I can see this being pretty fun for interactive stories. You could probably take the same world in completely different directions. How much control does the user have?

Maker
Exactly, that’s one of the fun parts. You can move around with WASD and prompt changes as the world evolves, so the same starting point can go in pretty different directions. We’re definitely interested in pushing the control further too. What kind of story would you try first?

Really fun demo. I spent way longer than expected just testing random prompts 😅

Maker

 Glad to hear you already have fun with R2🫶

​i have been looking for an engine that handles interactive storytelling without breaking context and this looks super impressive.

Maker
Give it a try plz！！🙌🏻

Generating continuous audio visual worlds instead of fixed clips is a massive step forward for AI video.

Maker
Sure thing. Continuous generation feels like a very different experience from stitching together fixed clips. Once the world can keep going and react to you, it starts to feel less like video generation and more like something you can actually inhabit lol

​i have been pretty skeptical about real time world models lately but seeing input actions dynamically alter the audio visual stream actually made a lot of sense.

Maker
Totally get the skepticism, that’s actually why we wanted R2 to feel interactive instead of just being another world model demo. Once your actions start changing the audiovisual stream in real time, it clicks a lot faster. Glad that part came through!

How much of that branching behavior is controlled by the creator versus generated by R2?

Maker
That’s actually a super interesting point. There’s always a bit of unpredictability with generative models, so how much comes from creator control vs. the model just taking things in its own direction is a really fun question. Hadn’t thought about it that way before.🤔
12
Next