Fei-Fei Li’s World Labs speeds up the world model race with its first commercial product, Marble

World Labs, the startup founded by AI pioneer Fei-Fei Li, is launching its first commercial world model product. marble Now available through freemium and paid tiers that let users turn text prompts, photos, videos, 3D layouts or panoramas into editable, downloadable 3D environments.

Launch of the Generative World Model, First released in limited beta The preview, from two months ago, comes a little more than a year after World Labs came out of stealth with $230 million in funding, and puts the startup ahead of competitors building world models. World models are AI systems that generate an internal representation of the environment, and can be used to predict future outcomes and plan actions.

Startups like Descartes and Odyssey have released free demos, and Google’s Genie is still in limited research preview. Marble differs from these – and even World Labs’ own real-time model, rtfm – Because it creates persistent, downloadable 3D environments instead of generating a world on the fly as you discover it. The company says this results in less morphing or inconsistency, and lets users export worlds as Gaussian splats, meshes, or video.

Marble is the first model of its kind to offer AI-native editing tools and a hybrid 3D editor that lets users block out spatial structures before AI fills in visual details.

“This is a whole new class of models that are creating 3D worlds, and it’s something that’s only going to get better over time. It’s something we’ve already improved a lot,” World Labs co-founder Justin Johnson told TechCrunch.

Last December, World Labs showed how its early models could generate interactive 3D scenes based on a single image. While impressive, the somewhat cartoonish scenes were not fully explorable as the activities were limited to a small area, and there were sometimes errors in presentation.

In my testing of the beta preview, I found that Marble created impressive worlds from image cues alone – from game-like environments to photorealistic versions of my living room. The visuals have been transformed around the edges, although this has clearly been improved in today’s launch. That said, the world I created in the beta using a single prompt looked better and more closely matches my intent now than with the same prompt.

I haven’t tested the editing features yet, though Johnson says they make Marble practical for near-term gaming, VFX, and virtual reality (VR) projects.

“One of our main themes for Marble moving forward is creative control,” Johnson said. “There should always be a quick way to generate something, but you also need to be able to go deeper and have a lot more control over the things you’re generating. You don’t want the machine to just take the wheel and pull all the creativity away from you.”

Marble’s impact on creative control starts with input flexibility. The beta only accepted single images, forcing the model to invent unseen details for the 360-degree view. With full launch, users can now upload multiple images or short clips to show a location from different angles and the model can generate fairly realistic digital twins.

Then we have Chisel, an experimental 3D editor that lets users block out rough spatial layouts (think walls, boxes, or planes) and then add text cues to dictate the visual style. Marble generates the world, separating structure from style – in the same way as HTML provides the structure of a website and CSS adds color. Unlike text-based editing, Chisel lets you manipulate objects directly.

“I can just go in there and grab the 3D block representing the couch and move it somewhere else,” Johnson said.

Another new feature that gives you more editing control is the ability to expand the world.

“Once you create a world, you can expand it one more time,” Johnson said. “When you go to a part of the world that starts to break up, you can basically tell the model to expand there or generate more worlds around where you currently are, and then it can add more detail to that area.”

Users who want to create a much larger space can combine multiple worlds with “Composer Mode”. Johnson demonstrated this for me with two worlds he had already created – a room made of cheese with grapevine chairs, and a futuristic living room in space.

path to spatial intelligence

Marble is available through four subscription tiers: Free (four generations from text, image, or panorama), Standard ($20/month, 12 generations plus multi-image/video input and advanced editing), Pro ($35/month, 25 generations with visual extensions and commercial rights), and Max ($95/month, all features and 75 generations).

Johnson thinks the initial use cases for Marble will be gaming, visual effects for film, and virtual reality.

Game developers have mixed feelings about the technology. a fresh Game Developers Conference Survey A third of respondents believe generative AI has a negative impact on the games industry – 12% more than the survey reported a year ago. Intellectual property theft, energy consumption and reduction in quality from AI-generated content were among the top concerns aired. And last year, a wired The investigation found that game studios like Activision Blizzard are using AI to cut corners and deal with degradation.

In gaming, Johnson sees developers using Marble to generate background environments and ambient spaces and then importing those assets into a game engine like Unity or Unreal Engine to add interactive elements, logic, and code.

“It’s not designed to replace the entire existing pipeline for gaming, but rather just to give you properties that you can drop into that pipeline,” he said.

For VFX work, according to Johnson, Marble AI bypasses the inconsistencies and poor camera controls that plague the video generator. Its 3D assets let artists stage scenes and control camera movements with frame-perfect precision, he said.

While Johnson said World Labs is not focusing on virtual reality (VR) applications right now, he said the industry is “hungry for content” and excited About the launch. Marble is already compatible with the Vision Pro and Quest 3 VR headsets, and each generated world can be viewed in VR today.

There may also be potential use cases for marble for robotics. Johnson said that unlike image and video generation, robotics does not have the benefit of large stores of training data. But with a generator like Marble, it becomes easier to simulate the training environment.

according to a recent manifesto According to Fei-Fei Li, CEO and co-founder of World Labs, Marble represents the first step toward creating a “truly spatially intelligent world model.”

Lee believes that “the next generation of world models will enable machines to achieve an entirely new level of spatial intelligence.” If large language models can teach machines to read and write, Li hopes systems like Marble can teach them to see and build. She says the ability to understand how things exist and interact in three-dimensional spaces could eventually help machines achieve breakthroughs beyond gaming and robotics and even into science and medicine.

“Our dream of truly intelligent machines will not come true without spatial intelligence,” Lee wrote.

Got any sensitive information or confidential documents? We’re reporting on the inner workings of the AI ​​industry – from the companies shaping its future to the people affected by their decisions. Contact Rebecca Bellan rebecca.bellan@techcrunch.com or Russell Brandom at russell.brandom@techcrunch.com. For secure communication, you can contact her via Signal at @rebeccabellan.491 and russellbrandam.49.