World Labs (@theworldlabs) / X

World Labs

480 posts

World Labs

@theworldlabs

World Labs is a spatial intelligence company, building frontier models that can perceive, generate, and interact with the 3D world.

San Francisco, CA

Joined January 2024

Pinned
World Labs
@theworldlabs
Apr 7
We're excited to be rolling out two model updates today! Marble 1.1: Improves lighting and contrast, with a major reduction in visual artifacts. Marble 1.1-Plus: Our new model built for scale. Create larger, more complex environments than ever before.
00:00
213K
World Labs
@theworldlabs
Jun 18
From Marble to Unreal Engine. Learn how to bring Marble-generated splats into @UnrealEngine using the VIVE Mars Nova 3DGS Plugin, including collision setup, environment lighting, and scene configuration. Watch here ↓
00:00
14K
World Labs
@theworldlabs
Jun 18
Created for the World Labs community by @cglewis92.
2.6K
World Labs reposted
Brittani Natali
@BrittaniNatali
Jun 16
One of my favorite things about gaussian splats is how quickly a few images can become something you can explore. This environment was stitched together from three images using @theworldlabs Breakdown below ↓
00:00
12K
World Labs
@theworldlabs
Jun 15
The future of AI should be grounded in human agency, creativity, and understanding. @FastCompany explores the rise of world models and features insights from World Labs cofounder @drfeifei: "Her vision for World Labs—and its human-centered future—is both consistent and
Fast Company
@FastCompany
Jun 15
“I’m keenly aware of multiple clocks,” Li says. “They’re all ticking.” f-st.co/oOa5Sp3
54K
World Labs
@theworldlabs
Jun 12
Today we are sharing three new research papers, each exploring a new way to generate 3D content by leveraging large-scale generative models and 2D priors. These projects were led by our incredible interns @HaoZhang623 @BDuisterhof @DrTunnels [1/4]
00:00
109K
World Labs
@theworldlabs
Jun 12
Replying to @theworldlabs
Modality Forcing adapts text-to-image models to reason jointly about text, images, and depth. It shows that text-to-image is a scalable pretraining objective for 3D reasoning, and how text-to-RGBD, depth estimation, and depth-to-image can be unified in a single model. [3/4]
Bardienus Duisterhof
@BDuisterhof
Jun 12
Introducing Modality Forcing, a recipe for post-training T2I models for SOTA RGB-Depth generation! Text-to-image (T2I) models learn rich representations of the spatial world. How do we build on this prior for high-quality depth generation? modality-forcing.github.io 🧵 [1/6]
00:00
7K
World Labs
@theworldlabs
Jun 12
Flex4DHuman lifts monocular video into dynamic 4D Gaussians. A video diffusion model is finetuned to generate synchronized multiview videos which are distilled into 4D Gaussians. With this method, a video of a person dancing can be lifted to 4D and composted into a 3D world.
Andy Cheng
@DrTunnels
Jun 12
Introducing Flex4DHuman: a video diffusion model that turns a single video or sparse camera views into synchronized multi-view videos using camera-pose and text conditioning. The generated views can be reconstructed into 4D Gaussian splats for AR/VR, gaming, and video
00:00
4.9K
World Labs
@theworldlabs
Jun 10
Millions of Gaussian splats, rendered directly in the browser. For ICARE, the creative team designed and iterated 7 worlds with Marble, then streamed them in real time using Spark 2.0 and Three.js Here's how the pipeline works ↓🧵
00:00
12K
World Labs
@theworldlabs
Jun 10
Replying to @theworldlabs
One of the more interesting parts of the pipeline was the trailer itself. Rather than rebuilding scenes for a separate cinematic workflow, the team connected Blender directly to the live Spark 2.0 runtime and animated cameras in the same worlds used for gameplay. more in the
World Labs
@theworldlabs
Jun 9
We turned dreams into worlds. Then filled them with history's greatest minds. Not a video. A world, running directly in your browser. Step inside ↓
00:00
3.6K
World Labs
@theworldlabs
Jun 10
ICARE is an early example of what's possible when world models become part of the creative workflow, helping teams rapidly create, iterate on, and connect interactive experiences. Play it now:
Icare
From icare-game.world
1.5K
World Labs
@theworldlabs
Jun 9
We turned dreams into worlds. Then filled them with history's greatest minds. Not a video. A world, running directly in your browser. Step inside ↓
00:00
68K
World Labs
@theworldlabs
Jun 9
Replying to @theworldlabs
A glimpse behind the scenes. The launch video wasn't rendered separately. The team connected Blender to the live Spark 2.0 runtime, animating cameras and previewing the splat-rendered result in real time. The same Marble gameplay worlds also helped create the launch video.
00:00
2.3K
World Labs
@theworldlabs
Jun 9
Later this week we'll share a deeper look at the workflows behind ICARE: • World generation and design • Browser-native splat streaming • Gameplay and interaction systems • AI-assisted production Marble • @sparkjsdev • @threejs • @Blender
00:00
2.1K
World Labs reposted
a16z
@a16z
Jun 8
World Labs CEO Dr. Fei-Fei Li says AI must change how we teach and evaluate students: "AI must change learning. AI must change K-16 learning." "The most precious resource of our entire world is human capital." "When we have gotten the technology that can answer standardized
00:00
81K
World Labs reposted
Fei-Fei Li
@drfeifei
Jun 3
Article
A Functional Taxonomy of World Models
“The world is everything that is the case.” — Ludwig Wittgenstein, Tractatus Logico-Philosophicus, 1921 The world is not made of words. In an earlier essay, we argued that spatial intelligence is AI’s...
1M