Editorial illustration for World Labs Launches Marble 3D Multimodal Model and Creator Workspace
Marble 3D Model Disrupts AI Spatial Computing Landscape
World Labs Unveils Marble 3D Multimodal Model and Marble Labs Workspace
World Labs just made the generative 3D space a bit more real. It’s a field thick with research demos and vaporware, but their new Marble model and the accompanying Marble Labs workspace look like an actual tool you might use.
This isn't another speculative paper. They've built a platform for generating and editing 3D worlds, and they're releasing it for public use. The goal seems to be moving from impressive tech demos to a messy, usable creative process.
The company is betting that artists and engineers don't just want a magic button. They need a workshop.
Alongside the launch, World Labs introduced Marble Labs, a workspace for creators to explore workflows, case studies, and documentation. "It is where artists, engineers, and designers push the boundaries of world models," the company said. There are many other startups like Decart and Odyssey building world models and putting out free demos, and Google's Genie is still only in a research preview.
Marble supports text-to-world and image-to-world generation, as well as multi-image and video inputs for greater control over scene structure. Worlds can be exported as Gaussian splats, triangle meshes, collider meshes, or videos with pixel-level camera control. Marble now includes AI-native world editing tools, enabling object removal, style changes, and structural modifications.
World Labs also unveiled Chisel, an experimental 3D sculpting mode that lets users design coarse layouts and apply styles through prompts. "Chisel decouples structure from style," according to the company. Users can expand worlds through one-step enlargement or compose multiple worlds to build large spaces.
Marble's direct competition is Google's locked-down Genie and a handful of other startups. Its practical edge might be the export options and the editing suite. You can generate a world, then actually change it.
Remove a tree. Shift the style. That moves it closer to a production asset than a one-off render.
Chisel is the more curious experiment. It lets you rough out a 3D shape, then prompt a style onto it. This is a different approach, prioritizing structural control over pure textual whimsy.
The real test is whether any of this sticks. Can it handle the specific, weird demands of a real project? The workspace suggests World Labs knows the answer lies with users, not in another press release.
They've provided the tools. Now someone has to go make something ugly and useful with them.
Related Reading
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
From Archives Reverie Unveils India-Focused STT Model, 1.5× Faster Than Deepgram
From Archives Anthropic to Spend USD 50 B on US Data Centres, Adding 2,400 Jobs
Common Questions Answered
What unique capabilities does the Marble 3D multimodal model offer for creators?
The Marble 3D model supports advanced text-to-world and image-to-world generation techniques, allowing creators to transform various input types into spatial computing environments. This capability represents a significant advancement in generative AI, enabling more flexible and intuitive content creation across multiple input formats.
How does Marble Labs differentiate itself from other world model platforms?
Marble Labs provides a dedicated workspace for artists, engineers, and designers to explore world model workflows, documentation, and case studies. Unlike competitors like Decart or Odyssey who offer limited demos, World Labs is building a comprehensive ecosystem that bridges technical and creative boundaries in multimodal AI generation.
What input types can creators use with the Marble 3D model?
The Marble 3D model supports multiple input types, including text, single images, multi-image sequences, and video inputs for world generation. This versatility allows creators to transform diverse source materials into complex spatial computing environments, expanding the potential for AI-driven creative workflows.
Further Reading
- Marble: A Multimodal World Model - World Labs — World Labs
- Generating Bigger and Better Worlds - World Labs — World Labs
- Research & Insights - World Labs — World Labs
