Loading video...

Video Failed to Load

Go Home

🚀🎬We introduce TMD (Transition Matching Distillation): 480p videos generated from text prompts in < 3 NFEs! 1️⃣Main backbone for feature extraction and lightweight head for iterative refinement 2️⃣Distilled from Wan2.1 14B T2V combining MeanFlow & DMD2 🔗

julberner's profile picture

Julius Berner

1,174 subscribers

16,857 views • 5 months 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

Nano Banana 2 is the "all-in-one" for visual prototyping. ⚡️ By combining character consistency, advanced world knowledge, and precision text, it’s built for fast workflows. From iterative character design to beautiful mockups, try it out today in Antigravity.
0:57

Nano Banana 2 is the "all-in-one" for visual prototyping. ⚡️ By combining character consistency, advanced world knowledge, and precision text, it’s built for fast workflows. From iterative character design to beautiful mockups, try it out today in Antigravity.

Google Antigravity

90,638 views • 4 months ago

Twitch streamers are cooked. FlowAct-R1 by ByteDance: - Interactive humanoid video generation; - streams infinite-length 480p 25fps, ~1.5s TTFF; - one-shot full-body control; - based on Seedance; - distillation cuts inference to just 3 NFEs.
1:26

Twitch streamers are cooked. FlowAct-R1 by ByteDance: - Interactive humanoid video generation; - streams infinite-length 480p 25fps, ~1.5s TTFF; - one-shot full-body control; - based on Seedance; - distillation cuts inference to just 3 NFEs.

Wildminder

33,598 views • 5 months ago

Generate photorealistic textures from text prompts with the new Text to Texture feature, now available in #Substance3DSampler (beta). Improve your workflow by spending less time hunting for reference imagery, and more time bringing your creative vision to life.
1:04

Generate photorealistic textures from text prompts with the new Text to Texture feature, now available in #Substance3DSampler (beta). Improve your workflow by spending less time hunting for reference imagery, and more time bringing your creative vision to life.

Adobe Substance 3D

484,254 views • 2 years ago

WanVideo sliding context window test on the 480p 14B T2V -model, worked surprisingly well but incredibly slow (50 mins on a 5090, 513 frames at 832x480)
0:32

WanVideo sliding context window test on the 480p 14B T2V -model, worked surprisingly well but incredibly slow (50 mins on a 5090, 513 frames at 832x480)

Jukka Seppänen

13,085 views • 1 year ago

holy moly 🔥 Pusa 1.0 is a new image-to-video finetune based on Wan2.1-T2V-14B that surpasses Wan-I2V-14B and is: > 200x cheaper to train (only cost $500 🤯) > 5x faster inference > supports start-end frames, video extension and video completion
0:35

holy moly 🔥 Pusa 1.0 is a new image-to-video finetune based on Wan2.1-T2V-14B that surpasses Wan-I2V-14B and is: > 200x cheaper to train (only cost $500 🤯) > 5x faster inference > supports start-end frames, video extension and video completion

Dreaming Tulpa 🥓👑

15,688 views • 11 months ago

Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery TL;DR: Skyfall-GS converts satellite images to explorable 3D urban scenes using diffusion models, with real-time rendering performance. Contributions: • We introduce Skyfall-GS, the first method to synthesize immersive, real-time, free-flight navigable 3D urban scenes solely from multi-view satellite imagery using generative refinement. • An open-domain refinement approach leverages pre-trained text-to-image diffusion models without domain-specific training. • A curriculum-learning-based iterative refinement strategy progressively enhances reconstruction quality from higher to lower viewpoints, significantly improving visual fidelity in occluded areas.
1:05

Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery TL;DR: Skyfall-GS converts satellite images to explorable 3D urban scenes using diffusion models, with real-time rendering performance. Contributions: • We introduce Skyfall-GS, the first method to synthesize immersive, real-time, free-flight navigable 3D urban scenes solely from multi-view satellite imagery using generative refinement. • An open-domain refinement approach leverages pre-trained text-to-image diffusion models without domain-specific training. • A curriculum-learning-based iterative refinement strategy progressively enhances reconstruction quality from higher to lower viewpoints, significantly improving visual fidelity in occluded areas.

MrNeRF

66,111 views • 8 months ago

From text to reality: MIT researchers find new ways to use LLMs to help in design & manufacturing. These models can convert text prompts to CAD, generate manufacturing instructions, and search for optimal designs:
1:25

From text to reality: MIT researchers find new ways to use LLMs to help in design & manufacturing. These models can convert text prompts to CAD, generate manufacturing instructions, and search for optimal designs:

MIT CSAIL

59,152 views • 2 years ago

(1/2) LightIt: Illumination Modeling and Control for Diffusion Models! #CVPR2024 We facilitate lighting control for novel image generation from text prompts. We can also edit lighting for a given input image. Video: Project:
0:31

(1/2) LightIt: Illumination Modeling and Control for Diffusion Models! #CVPR2024 We facilitate lighting control for novel image generation from text prompts. We can also edit lighting for a given input image. Video: Project:

Matthias Niessner

19,849 views • 2 years ago

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.
0:22

TurboEdit Instant text-based image editing discuss: We address the challenges of precise image inversion and disentangled image editing in the context of few-step diffusion models. We introduce an encoder based iterative inversion technique. The inversion network is conditioned on the input image and the reconstructed image from the previous step, allowing for correction of the next reconstruction towards the input image. We demonstrate that disentangled controls can be easily achieved in the few-step diffusion model by conditioning on an (automatically generated) detailed text prompt. To manipulate the inverted image, we freeze the noise maps and modify one attribute in the text prompt (either manually or via instruction based editing driven by an LLM), resulting in the generation of a new image similar to the input image with only one attribute changed. It can further control the editing strength and accept instructive text prompt. Our approach facilitates realistic text-guided image edits in real-time, requiring only 8 number of functional evaluations (NFEs) in inversion (one-time cost) and 4 NFEs per edit. Our method is not only fast, but also significantly outperforms state-of-the-art multi-step diffusion editing techniques.

AK

16,062 views • 1 year ago

🚀 Meet MassGen! 🛠️ An open-source project for multi-agent scaling. Inspired by Grok Heavy & Gemini DeepThink. Enable parallel intelligence sharing, iterative refinement & consensus across agents. Google AI OpenAI xAI MVP out now—star & feedback! 👇
1:03

🚀 Meet MassGen! 🛠️ An open-source project for multi-agent scaling. Inspired by Grok Heavy & Gemini DeepThink. Enable parallel intelligence sharing, iterative refinement & consensus across agents. Google AI OpenAI xAI MVP out now—star & feedback! 👇

Chi Wang

17,156 views • 11 months ago

We just solved creating infinitely long AI videos from a single prompt. We built Latted as the unified workspace for AI video projects. Then, we made the world's first AI generated feature film with it!
5:18

We just solved creating infinitely long AI videos from a single prompt. We built Latted as the unified workspace for AI video projects. Then, we made the world's first AI generated feature film with it!

emre

13,930 views • 4 months ago

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.
1:20

Presto! Distilling Steps and Layers for Accelerating Music Generation Despite advances in diffusion-based text-to-music (TTM) methods, efficient, high-quality generation remains a challenge. We introduce Presto!, an approach to inference acceleration for score-based diffusion transformers via reducing both sampling steps and cost per step. To reduce steps, we develop a new score-based distribution matching distillation (DMD) method for the EDM-family of diffusion models, the first GAN-based distillation method for TTM. To reduce the cost per step, we develop a simple, but powerful improvement to a recent layer distillation method that improves learning via better preserving hidden state variance. Finally, we combine our step and layer distillation methods together for a dual-faceted approach. We evaluate our step and layer distillation methods independently and show each yield best-in-class performance. Our combined distillation method can generate high-quality outputs with improved diversity, accelerating our base model by 10-18x (230/435ms latency for 32 second mono/stereo 44.1kHz, 15x faster than comparable SOTA) -- the fastest high-quality TTM to our knowledge.

AK

30,430 views • 1 year ago

VSTAR Generative Temporal Nursing for Longer Dynamic Video Synthesis Despite tremendous progress in the field of text-to-video (T2V) synthesis, open-sourced T2V diffusion models struggle to generate longer videos with dynamically varying and evolving content. They tend to
0:45

VSTAR Generative Temporal Nursing for Longer Dynamic Video Synthesis Despite tremendous progress in the field of text-to-video (T2V) synthesis, open-sourced T2V diffusion models struggle to generate longer videos with dynamically varying and evolving content. They tend to

AK

36,982 views • 2 years ago

Wan2.1-VACE 14B & 1.3B are now natively supported in ComfyUI! This model from Wan brings all-in-one editing capability to your video generation: 🔹Text-to-Video & Image-to-Video 🔹 Video-to-video (Pose & depth control) 🔹 Inpainting & Outpainting 🔹 Character + object reference
0:36

Wan2.1-VACE 14B & 1.3B are now natively supported in ComfyUI! This model from Wan brings all-in-one editing capability to your video generation: 🔹Text-to-Video & Image-to-Video 🔹 Video-to-video (Pose & depth control) 🔹 Inpainting & Outpainting 🔹 Character + object reference

ComfyUI

21,517 views • 1 year ago