• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    World Labs Releases Atlas Spatial Foundation Model

    Founding member Keunhong Park announces the in-house trained model at World Labs.

    KP
    1 Source, 29d ago, first seen 29d ago

    TLDR

    Keunhong Park, a founding member at World Labs, posted that Atlas is now live. He described it as a spatial foundation model trained entirely in-house from scratch. Park noted that after shipping RTFM the previous year, he believed a single camera-conditioned model could handle most spatial generation tasks, and Atlas represents that effort scaled up. The post links to a World Labs announcement and includes a generated headline mentioning a pixel-based launch.

    Combined views

    23.6K

    1 Source, first seen 29d ago

    Combined views

    23.6K

    1 Source, first seen 29d ago

    104 likes
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    104 likes
    1 comments
    29 saves
    12 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    1 comments
    29 saves
    12 reposts

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    1 Source

    @KeunhongPAtlas is live! A spatial foundation model, trained in-house from scratch. After shipping RTFM last year, I was convinced a single camera-conditioned model could unify most spatial generation tasks. Atlas is that bet at scale. The core idea👇

    1 Source

    @KeunhongPAtlas is live! A spatial foundation model, trained in-house from scratch. After shipping RTFM last year, I was convinced a single camera-conditioned model could unify most spatial generation tasks. Atlas is that bet at scale. The core idea👇