Fei-Fei Li’s World Labs accelerates the global model race with Marble, its first commercial product

World Labs, the startup founded by artificial intelligence pioneer Fei-Fei Li, is launching its first commercial model globally. Marble is now available through freemium and paid tiers that allow users to turn their text messages, photos, videos, 3D layouts or panoramas into editable, downloadable 3D environments.

The launch of the productive global model, first released in limited beta The preview 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 that build global models. Global models are artificial intelligence systems that create an internal representation of an environment and can be used to predict future outcomes and plan actions.

Startups like Decart and Odyssey have released free demos, and Google’s Genie is still in limited research preview. Marble differs from them — even from World Labs’ real-time model, RTFM — because it creates permanent downloadable 3D environments instead of creating worlds on the fly as you explore. This, the company says, results in less formatting or inconsistency and allows users to export worlds as Gaussian splats, meshes or video.

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

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

Last December, World Labs showed how its early models could create interactive 3D scenes based on a single image. While impressive, the somewhat cartoonish scenes weren’t fully explorable as movements were limited to a small area and there were occasional rendering errors.

In testing the beta preview, I found that Marble created impressive worlds from just image prompts — from game-like environments to photorealistic versions of my living room. Scenes were morphed at the edges, though apparently improved in today’s release. That being said, a world I had created in the beta using a single prompt looked better and fit my intent better than the same prompt now.

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

“One of our main themes for Marble is creative control,” Johnson said. “There should always be a fast track to creating something, but you should be able to dive even deeper and have a lot of control over the things you create. You don’t want the machine to just take the wheel and pull all that creativity away from you.”

Marble’s creative control begins with input flexibility. The beta only accepted single images, forcing the model to invent invisible details for a 360-degree view. With the full launch, users can now upload multiple images or short clips to show a space from different angles and have the model create 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 messages to guide the visual style. Marble creates the world by decoupling structure from style — similar to how 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 that represents the couch and move it somewhere else,” Johnson said.

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

“Once you create a world, you can expand it up to once,” Johnson said. “When you move to a part of the world that’s starting to break up, you can basically tell the model to expand there, or create more world near where you’re currently at, and then add more detail to that area.”

Users who want to create extremely large spaces can combine multiple worlds with the “composite mode”. Johnson demonstrated this for me with two worlds he had already built – a cheese room with grape chairs and another of a futuristic meeting room in space.

The Path to Spatial Intelligence

Marble is available through four subscription levels: Free (four generations of text, image or panorama), Standard ($20/month, 12 generations plus multi-image/video import and advanced editing), Pro ($35/month, 25 generations with scene extension and commercial rights), and Max ($95/month, all features and generations).

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

Game developers have mixed feelings about the technology. A recent one Game Developers Conference Research found that a third of respondents believed that genetic AI is having a negative impact on the games industry – 12% more than the survey showed last year. Intellectual property theft, power consumption and quality reduction from AI-generated content were among the top concerns raised. And last year, a Wired The research found that game studios like Activision Blizzard are using artificial intelligence to cut corners and fight attrition.

In gaming, Johnson sees developers using Marble to create background environments and ambient spaces, then importing those elements into game engines 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 to just give you assets that you can drop into that pipeline,” he said.

For VFX work, Marble bypasses the inconsistency and poor camera control that plague AI video generators, according to Johnson. Its 3D elements allow artists to stage scenes and control camera movements with absolute frame accuracy, he said.

While Johnson said World Labs isn’t focusing on virtual reality (VR) applications right now, he noted that the industry is “hungry for content” and irritated about the launch. Marble is already compatible with the Vision Pro and Quest 3 VR headsets, and every world created can be viewed in VR today.

Marble may also have potential use cases for robotics. Johnson noted that unlike image and video generation, robotics does not have the benefit of a large storehouse of training data. But with generators like Marble, it becomes easier to simulate educational environments.

According to a recent manifesto from 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 global models will enable machines to achieve spatial intelligence at a whole new level.” If large language models can teach machines to read and write, Li hopes that systems like Marble can teach them to see and construct. He says the ability to understand how things exist and interact in three-dimensional spaces may eventually help machines make discoveries beyond toys and robotics and even into science and medicine.

“Our dreams of truly intelligent machines will not be complete without spatial intelligence,” Lee wrote.

Do you have a sensitive tip or confidential documents? We cover the inner workings of the AI ​​industry — from the companies shaping its future to the people affected by their decisions. Contact Rebecca Bellan at rebecca.bellan@techcrunch.com or Russell Brandom at russell.brandom@techcrunch.com. For secure communication, you can reach them via Signal at @rebeccabellan.491 and russellbrandom.49.