Introducing Atlas: A World Model with Pixel-Perfect Camera Control
gowthami_s · x · 2026-09-02
World Labs introduced Atlas, a versatile multi-task model. As a camera-conditioned visual generation model rooted in a simple autoregressive design, Atlas learns and performs various tasks efficiently. The author shared a fan trailer demonstrating Atlas's capabilities, highlighting its elegance and malleability in generating complex visuals.
Related event: World Labs Launches Atlas, an Omni World Model for Spatial Intelligence(64 posts)→
More from Multimodal
- Apple's immersive horror short Remnants wows Vision Pro users with 3D presence — Scobleizer · 2026-10-18
- Proposal: a community-built open dataset to fix broken video-editing LoRAs — Emergency-Shake4529 · 2026-10-18
- H3 360° Orbit Camera LoRA for MiniMax Open-Sourced, Generates Smooth 5.2s Shots — merica420_69 · 2026-10-18
- 'Tesla in Pakistan' AI video made with Seedance 2.5 goes viral — SimplyAnnisa · 2026-10-18
- Borges's nine stories turned into walkable 3D worlds with AI model Fable 5.1 — emollick · 2026-10-18
- Midjourney cozy couple photo: full prompt shared for killing the AI plastic look — tisch_eins · 2026-10-18