Fly around and explore multiple worlds in Apple Vision Pro. Stitched worlds made in WorldLabs. Congratulations on the launch World Labs Fei-Fei Li

Recreating Home Alone street in minutes. Made using WorldLabs, Viewed in Apple Vision Pro using AirVis

3D Scanning took a huge leap this year. Below scans took less than a minute each, using just a phone. Scanned and Viewed with AirVis App apps.apple.com/us/app/airvis/…

Recreating the world of Avatar in just a few minutes. Made using WorldLabs. Viewed in AirVis app Apple: apps.apple.com/us/app/airvis/… Android: play.google.com/store/apps/det… Quest: meta.com/experiences/ai…

Gaussian Splat captured in AirVis app in just a couple of minutes Credits: SFMike

apple.github.io/ml-sharp/ 3DGS from Single Image. Free open source project by Apple, great work guys.

TurboDiffusion: 100–205× faster video generation on a single RTX 5090 🚀 Only takes 1.8s to generate a high-quality 5-second video. The key to both high speed and high quality? 😍SageAttention + Sparse-Linear Attention (SLA) + rCM Github: github.com/thu-ml/TurboDi… Technical

Single image to Splat in just 2 seconds. Testing out in AirVis app. Apple ML team did an amazing job.

We’re back in a Flash ⚡ Gemini 3 Flash is our latest model with frontier intelligence built for lightning speed, and pushing the Pareto Frontier of performance and efficiency. It outperforms 2.5 Pro while being 3x faster at a fraction of the cost. With this release, Gemini 3’s

Time to make a new model. Coming soon to AirVis

Image to Immersive scene. Nvidia Lyra vs Apple ML Sharp ML Sharp took about a second, Lyra took about an hour but use Video diffusion model self distillation. Sharp have commercial and distribution licensing restrictions. Lyra is Apache 2.0 free to distribute, use, train and

Snow day, captured in AirVis app, in minutes.

This AI can make 1080p videos in seconds. HiStream speeds up Wan video generations by 107x 🤯 Code under review haonanqiu.com/projects/HiStr…

NOT using Apple ML Sharp. Image -> Immersive Scene in seconds. 1 Million Gaussian splats from a single image using DA3-BASE & Brush (Apache 2.0). Results are not as good as ML Sharp, really wanted to add it , but couldn’t because of licensing restrictions. Try it for free

NVIDIA Cosmos Reason 2 is here. 🥳 An open, highly accurate reasoning vision language model for physical AI, featuring: ✅ Improved spatio-temporal understanding and timestamp precision ✅ Flexible deployment with 2B and 8B model sizes ✅ Long-context reasoning with up to 256K

OneShot [1M Splats] vs Walkthrough[2M splats] OneShot is MonoView, uses DA3 - Brush pipeline. Walkthrough is MultiView, uses Colmap - gSplat pipeline. Credits: SFMike. AirVis app. apps.apple.com/us/app/airvis/…