Loading video...

Video Failed to Load

Go Home

Having a few hours this afternoon after #CVPR2026? Consider visiting Red Rocks 20 minutes outside of Denver! For some inspiration, here's a world processed by SpAItial AI Echo-2 HQ model from a single phone image! Link to the world in🧵

rmbrualla's profile picture

Ricardo Martin-Brualla

3,329 subscribers

11,381 views • 1 month ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

This AI workflow is wild I used GPT Image 2 to create a character sheet, then animated it with Seedance 2.0 in just a few minutes Prompt in thread 🧵
0:13

This AI workflow is wild I used GPT Image 2 to create a character sheet, then animated it with Seedance 2.0 in just a few minutes Prompt in thread 🧵

Kiki

150,279 views • 13 days ago

Echo-2 is a physically-grounded world model from which we can distill meshes, point clouds, or 3DGS scene representations. Directly usable in a myriad of downstream applications from gaming to training robots. Want to build your own world? Try it here:
0:14

Echo-2 is a physically-grounded world model from which we can distill meshes, point clouds, or 3DGS scene representations. Directly usable in a myriad of downstream applications from gaming to training robots. Want to build your own world? Try it here:

SpAItial AI

21,416 views • 3 months ago

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.
0:18

Selected as a best paper finalist at #CVPR2026: PixelDiT from NVIDIA Research In most image generation models, a pretrained autoencoder compresses the image before any diffusion happens, causing quality loss that accumulates across the entire pipeline. PixelDiT, or Pixel Diffusion Transformers, removes this step entirely. It's a single-stage model that learns the diffusion process directly in pixel space, end-to-end.

NVIDIA AI

28,129 views • 1 month ago

World Models are heating up as Prof. Fei-fei Li Fei-Fei Li's World Labs World Labs ignites the space with their first product Marble 🔥🌍 Riding that momentum, we’re launching a World Model demo on Theta Network EdgeCloud, powered by Tencent Tencent 腾讯's open-source Hunyuan 3D World Model, and packaged as a standard model template for creators, devs, and AI worldbuilders. Text/Image → explorable 3D worlds. Single image → full geometry. Exportable, editable, fast! Try it now:
0:11

World Models are heating up as Prof. Fei-fei Li Fei-Fei Li's World Labs World Labs ignites the space with their first product Marble 🔥🌍 Riding that momentum, we’re launching a World Model demo on Theta Network EdgeCloud, powered by Tencent Tencent 腾讯's open-source Hunyuan 3D World Model, and packaged as a standard model template for creators, devs, and AI worldbuilders. Text/Image → explorable 3D worlds. Single image → full geometry. Exportable, editable, fast! Try it now:

Jieyi Long | Theta Network

25,503 views • 8 months ago

AI anime production is crazy now I used GPT Image 2 to create my first pirate character sheet, then animated the story with Seedance 2.0 in just a few minutes. As a longtime One Piece fan, this was way too much fun. Prompt in thread 🧵
0:13

AI anime production is crazy now I used GPT Image 2 to create my first pirate character sheet, then animated the story with Seedance 2.0 in just a few minutes. As a longtime One Piece fan, this was way too much fun. Prompt in thread 🧵

Kiki

12,387 views • 10 days ago

In 1983, there was an attempt to break the world record for fastest time to build a house from the ground up. The time to beat was 4 hours and 18 minutes, and they built the house in just over 2 hours and 45 minutes.
0:12

In 1983, there was an attempt to break the world record for fastest time to build a house from the ground up. The time to beat was 4 hours and 18 minutes, and they built the house in just over 2 hours and 45 minutes.

Historic Vids

1,470,641 views • 2 years ago

Will the world rely on a single #GenAI model? Don’t count on it. In this AI Academy episode, IBM Director of Research Darío Gil explores different approaches to #AI and how businesses can go from AI users to value creators. Watch it here:
0:15

Will the world rely on a single #GenAI model? Don’t count on it. In this AI Academy episode, IBM Director of Research Darío Gil explores different approaches to #AI and how businesses can go from AI users to value creators. Watch it here:

IBM

38,678 views • 2 years ago

This is a 🔥 launch for AI image generation! I uploaded four photos of myself to Krea and got a model that generates real-time images of me. Adjust the prompt and you'll immediately see a change - watch me travel around the world ✨
0:18

This is a 🔥 launch for AI image generation! I uploaded four photos of myself to Krea and got a model that generates real-time images of me. Adjust the prompt and you'll immediately see a change - watch me travel around the world ✨

Justine Moore

76,771 views • 1 year ago

Experimenting with image-to-world: creating consistent 3D environments from a single photo. As someone who studied architecture, this feels like a glimpse into the future of spatial creation and I’m excited to see where World Labs takes it.
0:17

Experimenting with image-to-world: creating consistent 3D environments from a single photo. As someone who studied architecture, this feels like a glimpse into the future of spatial creation and I’m excited to see where World Labs takes it.

Ian Curtis

61,738 views • 11 months ago

Electricity update: We've had no power for 4 hours in the morning, afternoon, and now again. That's a total of 12 hours daily. Internet is barely working. Everything's disrupted—from defrosted fridges in supermarkets to stalled manufacturing. 1/2 🧵
0:17

Electricity update: We've had no power for 4 hours in the morning, afternoon, and now again. That's a total of 12 hours daily. Internet is barely working. Everything's disrupted—from defrosted fridges in supermarkets to stalled manufacturing. 1/2 🧵

Maria Avdeeva

317,136 views • 2 years ago

Built a fully native, SwiftUI app for tracking our son’s naps and distributed it via TestFlight to my wife, over a few hours all from my phone(!!!) using Bitrig. It’s obviously not the best app in the world, but it’s fun and uses some iOS 26 APIs + modern components:
0:28

Built a fully native, SwiftUI app for tracking our son’s naps and distributed it via TestFlight to my wife, over a few hours all from my phone(!!!) using Bitrig. It’s obviously not the best app in the world, but it’s fun and uses some iOS 26 APIs + modern components:

Jordan Borth

18,718 views • 9 months ago

Thread on some of the most surreal & beautiful train rides in the world🧵 1. A ride from Bangalore to Idagunji, Karnataka, India 🇮🇳
0:18

Thread on some of the most surreal & beautiful train rides in the world🧵 1. A ride from Bangalore to Idagunji, Karnataka, India 🇮🇳

Raghu

1,614,153 views • 1 year ago

This update introduces a groundbreaking "Native Audio" capability. The model completely transforms the traditional AI video workflow of "first generating silent visuals, then manually adding voiceovers and sound effects." By deeply aligning the semantics of sounds and dynamic visuals from the physical world, VIDEO 2.6 enables the end-to-end generation of complete videos in a single go...
0:10

This update introduces a groundbreaking "Native Audio" capability. The model completely transforms the traditional AI video workflow of "first generating silent visuals, then manually adding voiceovers and sound effects." By deeply aligning the semantics of sounds and dynamic visuals from the physical world, VIDEO 2.6 enables the end-to-end generation of complete videos in a single go...

Angry Tom

76,743 views • 8 months ago

World Models are the path for some AI Models in the future. But how can we efficiently train these models to not only see the world the way humans do but to see the world in a new and unique way. By visualizing, what is normally sequenced audio patterns, we can derive much more insights. Here we see Paganini in a visual form that can than be described and transcribed into a World Model. We can observe connections in a manner that may not have been clear prior to the digitalization of music and sound in this way. The company with the most valuable potential in building a World Model is Tesla. Not that this type of visualization is being used, but that the mechanisms are in place, and the technology is in place for the company to thrive in this new form of AI.
0:19

World Models are the path for some AI Models in the future. But how can we efficiently train these models to not only see the world the way humans do but to see the world in a new and unique way. By visualizing, what is normally sequenced audio patterns, we can derive much more insights. Here we see Paganini in a visual form that can than be described and transcribed into a World Model. We can observe connections in a manner that may not have been clear prior to the digitalization of music and sound in this way. The company with the most valuable potential in building a World Model is Tesla. Not that this type of visualization is being used, but that the mechanisms are in place, and the technology is in place for the company to thrive in this new form of AI.

Brian Roemmele

57,454 views • 8 months ago

🎬 Hollywood better watch out. You only need a few minutes and a single AI model to create videos like this. Here’s the full workflow + prompt so you can try it too. 🔖 Bookmark for later
0:18

🎬 Hollywood better watch out. You only need a few minutes and a single AI model to create videos like this. Here’s the full workflow + prompt so you can try it too. 🔖 Bookmark for later

Pierrick Chevallier | IA

18,965 views • 1 year ago

This croissant is generated by a text-to-3d AI model. Just a short few years after Yakuza: Like a Dragon We've hit the pinnacle of bread-tech. More examples in reply + more info here:
0:15

This croissant is generated by a text-to-3d AI model. Just a short few years after Yakuza: Like a Dragon We've hit the pinnacle of bread-tech. More examples in reply + more info here:

Ben Ferns

124,961 views • 3 years ago

AI can now literally sell anything to anyone. Sora 2 is live inside arcads. just upload image, write a script, set the accent and tone. the output looks like a real person filmed it on their phone brands are still paying $500 for worse output than this. Here's how to do it 🧵
0:11

AI can now literally sell anything to anyone. Sora 2 is live inside arcads. just upload image, write a script, set the accent and tone. the output looks like a real person filmed it on their phone brands are still paying $500 for worse output than this. Here's how to do it 🧵

Farhan

23,675 views • 4 months ago

🧮Turn single image into a custom LoRA model in one click ! Should work for character and product ! This ComfyUI workflow: → Uses Gemini AI to generate 20 diverse prompts from your image → Creates 20 consistent variations with FLUX.1 Kontext → Automatically builds the dataset + trains the LoRA One image in → Trained LoRA out 🎯 #ComfyUI #LoRA #AIArt #FLUX #AutomatedAI ComfyUI BlackForestLabsAI - Unofficial 🔗 Check it out: This workflow was made for the hackathon organized by ComfyUI in SF yesterday
0:15

🧮Turn single image into a custom LoRA model in one click ! Should work for character and product ! This ComfyUI workflow: → Uses Gemini AI to generate 20 diverse prompts from your image → Creates 20 consistent variations with FLUX.1 Kontext → Automatically builds the dataset + trains the LoRA One image in → Trained LoRA out 🎯 #ComfyUI #LoRA #AIArt #FLUX #AutomatedAI ComfyUI BlackForestLabsAI - Unofficial 🔗 Check it out: This workflow was made for the hackathon organized by ComfyUI in SF yesterday

Lovis Odin

69,594 views • 1 year ago