3D scenes

Models 12

• Gen 3
  Project Eden

  By JetBrains

  Project Eden is VAST AI Research / Tripo’s research-preview world model for persistent multiplayer and multi-agent interaction inside a shared, consistent world state.

  🎮Interactive simulation🔀Synthetic data🎥Videos🎥3D scenes
  NewMultimodal

  Released 7d ago
• Gen 3
  TripoSplat

  By Tripo AI

  TripoSplat is TripoAI / VAST AI Research’s MIT-licensed image-to-3D model for converting a single 2D image into variable-count 3D Gaussians.

  🎲3D objects🔮Augmented reality🎮3D game assets🎥3D scenes
  NewMultimodal

  Released 17d ago
• Gen 3
  Echo 2

  By SpAItial

  Echo-2 is SpAItial’s frontier 3D world-generation model for creating physically grounded, explorable environments from text or image inputs. Instead of generating a video, it produces a persistent 3D-consistent scene representation that can be rendered in real time on any device, targeting digital twins, robotics, architecture, gaming, and immersive 3D creation.

  🎨Scene creation🧩Spatial intelligence🤖Robotics🏛Architectural visualization🎥3D scenes
  NewMultimodal

  Released 1mo ago
• Gen 3
  Odyssey 2 Max

  By New Odyssey

  Odyssey 2 Max appears to be Odyssey’s newest high-end AI video model tier, positioned above earlier Odyssey releases for stronger cinematic generation and editing. Public references point to it as a flagship model in the Odyssey video stack, but the exact official model page was not accessible enough to safely verify architecture or detailed specs.

  🚀Physics simulations🖥Computational system simulation🎥3D scenes
  NewMultimodal

  Released 1mo ago
• Gen 4
  MultiWorld

  By SReal AI

  MultiWorld is a scalable multi-agent, multi-view video world model for action-controllable simulation. It is designed to generate multi-view-consistent videos for multi-player games and multi-robot manipulation, while allowing the number of agents and camera views to scale flexibly.

  🎥Videos🤖Robotics🎥3D scenes
  NewVideo

  Released 1mo ago
• Gen 3
  HY World 2.0

  By Tencent

  HY-World 2.0 is Tencent Hunyuan’s multimodal 3D world model for reconstructing, generating, and simulating explorable worlds. It accepts text, single-view images, multi-view images, and videos, then outputs persistent 3D assets such as meshes and Gaussian splattings, aiming to create editable, navigable worlds rather than short-lived generated videos.

  🖌️3D🌎World creation📷3D scanning🎥3D scenes
  NewMultimodal

  Released 1mo ago
• Gen 4
  Waypoint 1.5

  By Overworld

  Waypoint-1.5 is Overworld’s real-time generative world model for interactive AI environments on consumer hardware. It is built to run locally with streaming performance, offering up to 720p at 60 FPS and introducing dual model tiers so generative worlds can run on a wider range of everyday GPUs.

  🌍Game worlds🎮Game design assistance🎥3D scenes
  NewVideo

  Released 1mo ago
• Gen 3
  Marble 1.1 Plus

  By World Labs

  Marble 1.1 Plus is World Labs’ most advanced model for creating the largest 3D worlds. It automatically expands its 3D spatial coverage when possible, aiming for bigger and more ambitious world generation than standard Marble 1.1.

  🖌️3D🖼️Image generation🎥3D scenes
  NewMultimodal

  Released 1mo ago
• Gen 3
  Marble 1.1

  By World Labs

  Marble 1.1 is World Labs’ fixed-cost world-generation model for creating higher-quality 3D worlds. It is the recommended starting model in the Marble lineup and is positioned as an improved-quality successor to Marble 1.0.

  🖌️3D🖼️Image generation🎥3D scenes
  NewMultimodal

  Released 2mo ago
• Gen 3
  Matrix Game 3.0

  By Skywork AI

  Matrix-Game 3.0 is Skywork’s open-source interactive world model for real-time long-form video generation. It is built to generate 720p streaming video conditioned on images, prompts, and actions, with long-horizon memory for consistency over extended sequences, and is optimized for interactive use with few-step sampling, quantization, and distillation.

  🎥Videos🎥Interactive videos🚀Physics simulations🎥3D scenes
  NewMultimodal

  Released 2mo ago
• Gen 3
  Asset Harvester

  By NVIDIA

  Asset Harvester is NVIDIA’s open framework for extracting reusable 3D assets from sparse autonomous-driving observations. It turns one or a few in-the-wild object views from driving logs into complete, simulation-ready 3D assets, even under occlusion, noisy calibration, and viewpoint bias.

  🎲3D objects🎥3D scenes🚗Autonomous driving
  NewMultimodal

  Released 2mo ago
• Gen 1
  InSpatio World

  By InSpatio

  InSpatio-World is a video-conditioned 4D world model that turns a reference video into a dynamic scene you can explore from new viewpoints through time.

  🎥3D videos🎥Spatial image to video🎥3D scenes
  New3d

  Released 2mo ago