Ruofan Liang retweeted
Kimi K3 combines strong 3D reasoning, coding, and vision capabilities to turn concepts, images, and videos into fully playable interactive experiences.
Kimi K3 achieves true "vision in the loop" by seamlessly iterating between code and live screenshots
Ruofan Liang retweeted
Can video generation models do for vision what LLMs did for language?
Introducing GenCeption from @GoogleDeepMind: one feed-forward video model for various vision tasks — SOTA, data-efficient, and emerging behaviors (ECCV 2026)
🌐 genception.github.io
(1/8)
I will also be there presenting LuxRemix!
Looking forward to your visit 🤗
BTW, I just released our massive 360 synthetic rendering dataset w/ light decomposition. Please check the thread.
Today, we’re presenting LuxRemix at #CVPR2026 — 10:45 poster session, poster #102!
Decompose indoor lighting into individual light sources from one image, then remix it (on/off, color, intensity) with real-time 3DGS relighting.
Come say hi 👋
luxremix.github.io
Ruofan Liang retweeted
World models are moving beyond offline generation towards interactive, real-time experiences.
Introducing ⚡FlashDreams⚡: an open-source high-performance inference and serving library built for autoregressive world models:
🔥 Up to 3.10× faster LingBot-World inference
🔥 Up to 2.12× faster Self-Forcing inference
🔥 Up to 1.40× faster Wan2.1 inference
🔥 8 integrated models
🔥 Multi-GPU, streaming, low-latency serving
🔥 Agentic skills that teach you how to use it
FlashDreams is designed for a new generation of AI systems that continuously evolve over time while responding to user interactions. It powers applications across robotics, autonomous vehicle simulation, gaming, and virtual worlds.
Github: github.com/NVIDIA/flashdream…
Docs: nvidia.github.io/flashdreams
Research page: research.nvidia.com/labs/sil…
Join the #flashdreams Discord channel at discord.gg/yTdHDqFP
FlashDreams is also the runtime backbone behind NVIDIA OmniDreams (github.com/nv-tlabs/omni-dre…)
1/n
#AI #WorldModels #FastInference #PhysicalAI #OpenSource #NVIDIA
Ruofan Liang retweeted
🔥 Check out our latest work!
🚀 Introducing Gamma-World: a generative multi-agent world model that scales beyond two players and supports real-time streaming at 24 FPS.
🔥 We release Gamma-World from @nvidia — a generative multi-agent world model that finally goes beyond 2 players.
⚡ 24 FPS real-time streaming
🧩 Simplex Rotary Agent Encoding — permutation-symmetric
🌐 Sparse Hub Attention: O(N²) → O(N)
🎯 2 → more players, zero-shot
🤖 Games → real multi-robot worlds
━━━━━━━━━━━━━━━
💥 THE SINGLE-AGENT ERA IS OVER. 💥
━━━━━━━━━━━━━━━
🔗 research.nvidia.com/labs/sil…
Fortunate to help @yxue_yxue on this project during our Meta internship 😃.
GeoRelight follows the research trajectory of DifffusionRenderer and UniRelight and delivers a significant quality improvement in relighting and geometry reconstruction for general human captures.
#CVPR2026 Highlight
How to make relighting more photorealistic? Make reconstruction happening together!
GeoRelight jointly resolves Geometry, Instrinsics, and Relighting, and proves they have mutual benefit (Geometry helps you know shadow and shading)
yuxuan-xue.com/georelight
Ruofan Liang retweeted
We open-sourced the code and model for UniRelight! 🎉
Given an input video and a target lighting configuration, our method jointly predicts a relit video and its corresponding albedo.
Code: github.com/nv-tlabs/UniRelig…
Model: huggingface.co/nvidia/UniRel…
The code is out for LuxDiT! 🎉
LuxDiT allows you to estimate high-quality HDR environment maps from images or videos. You can use our released Gradio web UI to quickly try your input.
🗒️Code: github.com/nv-tlabs/LuxDiT
💾Model: huggingface.co/nvidia/LuxDiT
💡 Introducing LuxDiT: a diffusion transformer (DiT) that estimates realistic scene lighting from a single image or video.
It produces accurate HDR environment maps, addressing a long-standing challenge in computer vision.
🔗Paper: arxiv.org/abs/2509.03680
Ruofan Liang retweeted
Spark 2.0 is here! 🚀
We’re redefining what’s possible on the web with a streamable LoD system for 3D Gaussian Splatting.
Built on Three.js, you can now stream massive 100M+ splat worlds to any device from mobile to VR using WebGL2. All open-source.
Dive into the tech 👇
Ruofan Liang retweeted
We scaled up Lyra to generate explorable 3D worlds! 🚀
Introducing Lyra 2.0 — turning a single image into a 3D world you can walk through, look back, and even drop a robot into 🤖
Code and Model available today!
🌐 Website: research.nvidia.com/labs/sil…
(1/N)
Ruofan Liang retweeted
My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept):
Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
I also took a picture of this classic place for the teaser demo of our #CVPR26 paper LuxRemix (luxremix.github.io/) 😄
I used to love to sit there on Saturday afternoon, hoping Einstein could give me some research ideas 🤣
Credits also to @c_richardt for composing the vid.
Ruofan Liang retweeted
Announcing NVIDIA DLSS 5, an AI-powered breakthrough in visual fidelity for games, coming this fall.
DLSS 5 infuses pixels with photorealistic lighting and materials, bridging the gap between rendering and reality.
Learn More → nvidia.com/en-us/geforce/new…
Ruofan Liang retweeted
📢Introducing 360Anything, our method for lifting any perspective image or video to gravity-aligned 360° panoramas without using any camera or 3D information. This enables consistent novel view synthesis and 3D scene reconstruction.
Project page: 360anything.github.io/
🧵
Excited to share our new work: LuxRemix ✨
We leverage diffusion models to decompose complex light transport into individual sources. You can interactively remix room lighting in 2D images or 3D GSplats 💡🎛️
Try the interactive light controls here: 🔗 luxremix.github.io
We’re excited to share LuxRemix: interactive light editing for indoor scenes! 🏠💡
Capture a room once, then turn individual lights on/off, change colors, and adjust intensity – all in real-time 3D from any viewpoint.
💡 luxremix.github.io/
📄 arxiv.org/abs/2601.15283
Just dropped a Blender-based data generation tool that can be used to render randomly composed synthetic scenes with all G-Buffer attributes. 😋
The generated data can be used for training generative rendering models such as our Diffusion Renderer, UniRelight, and LuxDiT.
🌐 Github Repo:
github.com/nexuslrf/composit…
