Generate an interactive 3D world ✨ I opened this project in Cursor and uploaded an image of a cozy bookstore. The agent then generated the environment, chose objects to turn into 3D meshes, added colliders, and created sound effects 🤯

This project uses @theworldlabs to generate the world from an input image. It selects items from the world that should be 3D objects, removes them from the image (using NB or GPT Image), and renders them with Hunyuan 3D. @ElevenLabs provides sound effects and ambient noise.

@ElevenLabs Huge props to @neilsonks for making this, he has a better explanation of how it works 👇 And also thanks to the guys at @fal for hosting almost every model so you only need two API keys for this entire thing!

You could basically easily create your own computer game!

Me throwing around the blob cat that the agent inexplicably rendered in 3D is already a game 😂

The crazy thing with Claude is that I created a software in basically 10 minutes to a task I always wanted to do in terms of data management. The key is not only to be very specific in your prompt, it's to ask the AI engine for 5-10 follow up questions to know your needs.

@Kyrannio Very cool

@ImVitek @SeKr0x @cryptosloth_nft

colliders from jpg?

this is hilarious

Image → 3D environment inside Cursor is a wild workflow. This is where AI starts acting like a production toolchain.

PSA: cloning the image-blaster repo doesn't make it free. You still need 2 paid API keys: • World Labs — 3D environments • — 3D models, SFX, image edits Each blast burns credit per generation.

i prefer codex so i tried making one:

This was just an Orchestrator, you would still need the api keys for this to work.. In other words, its not Claude, but just the usual services that does the work.

This is fantastic! If we require manual adjustments to turn 3D models with 4D effects, we can easily add joints.

Impressive should try this

Love this. The agent is not just making a 3D scene; it is deciding what becomes interactive, then wiring physics and sound. With SilicoVille, we are exploring the next layer: how does a world remember actions, costs, relationships, and consequences?

Are you going to do this for Codex too?

@theworldlabs May i know how much the cost almost? When you connect it with the two api ? Also is it require strong hardware or it’s ok with mac m2?

Cringe cursor yes Musk might buy it still say Anti Gravity is better yes @ me

love the idea

This is a powerful demonstration of AI turning simple images into rich interactive 3D worlds. Excellent progress.

the wild part isn't generating the bookstore. it's keeping the scene coherent after the fifth change request instead of collapsing into prompt soup.

This is a strong signal for how far agentic code generation has come—turning a single image into a full interactive environment with physics and audio is non-trivial. Curious about the mesh quality and performance at higher object counts.