🚀 Announcing Echo — our new frontier model for 3D world generation. Echo turns a simple text prompt or image into a fully explorable, 3D-consistent world. Instead of disconnected views, the result is a single, coherent spatial representation you can move through freely. This is part of a bigger... shift in AI: from generating pixels and tokens to generating spaces. Echo predicts a geometry-grounded 3D scene at metric scale, meaning every novel view, depth map, and interaction comes from the same underlying world — not independent hallucinations. Once generated, the world is interactive in real time. You control the camera, explore from any angle, and render instantly — even on low-end hardware, directly in the browser. High-quality 3D world exploration is no longer gated by expensive equipment. Under the hood, Echo infers a physically grounded 3D representation and converts it into a renderable format. For our web demo, we use 3D Gaussian Splatting (3DGS) for fast, GPU-friendly rendering — but the representation itself is flexible and can be easily adapted. Why this matters: consistent 3D worlds unlock real workflows — digital twins, 3D design, game environments, robotics simulation, and more. From a single photo or a line of text, Echo builds worlds that are reliable, editable, and spatially faithful. Echo also enables scene editing and restyling. Change materials, remove or add objects, explore design variations — all while preserving global 3D consistency. Editing no longer breaks the world. This is only the beginning. Echo is the foundation for future world models with dynamics, physical reasoning, and richer interaction — environments that don’t just look right, but behave right. Explore the generated worlds on our website and sign up for the closed beta. The era of spatial intelligence starts here. 🌍 #Echo #WorldModels #SpatialAI #3DFoundationModels Check it out:show more

I found the model amazing — it just needs to be enabled for viewing on VR devices.

Team, huge congrats for releasing the model only 6 months after the Seed round!! 2026 will be 🔥🔥

Can i wire these up in Unity?

Yes, the outputs are distilled into 3dgs representations and can easily be integrated!

Congrats!!!

Are you able to put people or characters into the space?

@Scobleizer Who owns the images/videos produced?

So excited for what's to come!

We'll change the world, obviously 🔥🔥🔥

Now if we were to integrate VR and Voice prompting then we're almost at the stage where you can walk into your world, point and change.

@Scobleizer I wrote a script 10 years ago about such world. One of the rules: no option to remove an item as it may impose a risk of harm to a user (real world). I would be ok to add anything to a specific not reachable areas. Some areas (interior/ exterior would consist only of dummies

Congrats!

Amazing ✨

Will definitely be checking this out!

Are the assets downloadable for modeling programs if edits are wanted?

3D world generation model yields coherent spatial representation from text.

This is incredible

@jon_barron WATTBA!

Can’t wait for my first braindance

Impressive!

Thanks!

Cool model, but you need a better viewer:

What other sources can be used? 360 Photos? Have played around with Gaussian Splats lately, VR interaction? three.js?

API soon?

Incredible! 👀

Absolutely incredible work, congratulations on getting all the way to this point - and can't wait to see how this improves! Can you combine splats with collisions (underlying mesh)?

This will be so useful for home remodelling services. Gives non visual people a way to actually see what the room would look like instead of just pics. Would it be possible to extract pics out of this model when flipped to a different angle/direction?

This level of detail from a single image prompt is impressive. The normal maps and textures look solid.

@levelsio is this something you're trying to set up too?

Hello. We've been experiencing high volumes of traffic, causing the site to slow down. My computer has low specs, so the sample file has stopped working and is now buried in the floor of the room. I hope that it will eventually run smoothly even on low-spec computers.

What's the maximum resolution you can output? I wonder if this could be used for various scenarios in a movie production.

Real time? On what hardware?

Waitlit?

please let me in!

That’s a lot of EM dashes 😂

Impressive!

That’s insane. I love it 😍