Загрузка видео...

Не удалось загрузить видео

На главную

freaking massive! LPM 1.0 -17B DiT for real-time, full-duplex conversational video generation. Avatars on huge steroids. - synchronized speaking, listening - 0.35s latency; - identity consistency for 10+ mins; - DMD distillation for 2-step online generation; - 480P/720P at 24 fps LiveAvatar, Kling-Avatar-2, OmniHuman -cooked.

wildmindai's profile picture

Wildminder

10,305 subscribers

19,038 просмотров • 2 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Avatar V API is live at $0.05/sec The highest-quality AI avatar model for developers Benchmarked against Veo 3.1, Kling O3 Pro, OmniHuman 1.5, and Seedance 2.0 on cross-scene talking-head generation Avatar V won every category Research report + API ↓
0:47

Avatar V API is live at $0.05/sec The highest-quality AI avatar model for developers Benchmarked against Veo 3.1, Kling O3 Pro, OmniHuman 1.5, and Seedance 2.0 on cross-scene talking-head generation Avatar V won every category Research report + API ↓

HeyGen

286,594 просмотров • 1 месяц назад

StreamDiT Real-Time Streaming Text-to-Video Generation StreamDiT enables real-time text-to-video generation at 16 FPS on a single GPU (H100)
1:00

StreamDiT Real-Time Streaming Text-to-Video Generation StreamDiT enables real-time text-to-video generation at 16 FPS on a single GPU (H100)

AK

43,659 просмотров • 11 месяцев назад

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 просмотров • 5 месяцев назад

1/ This Lemon Slice 2 deployment represents (afaik) the first production-scale real-time diffusion transformer for interactive video generation. Available as an API or embeddable widget. The only real-time avatar tool for any character, any style.
1:00

1/ This Lemon Slice 2 deployment represents (afaik) the first production-scale real-time diffusion transformer for interactive video generation. Available as an API or embeddable widget. The only real-time avatar tool for any character, any style.

Chubby♨️

15,868 просмотров • 6 месяцев назад

Introducing The Matrix --- a foundation world model for generating infinite-length, hyper-realistic videos with real-time, frame-level control: - Infinite-length video generation - 720p high-quality rendering - Real-time, frame-level control at 16 FPS - Generalization to real-world video control 🔗Blog: 📄Paper: 💻Code & Playable Demo: Coming soon! Key Innovation: A brand new technique called the shift-window denoise process model, enabling auto-regressive generation for diffusion and consistency models in real-time. Special thanks to project leader Ruili Feng and the entire Matrix team for their dedication and hard work over the year-long project.
1:00

Introducing The Matrix --- a foundation world model for generating infinite-length, hyper-realistic videos with real-time, frame-level control: - Infinite-length video generation - 720p high-quality rendering - Real-time, frame-level control at 16 FPS - Generalization to real-world video control 🔗Blog: 📄Paper: 💻Code & Playable Demo: Coming soon! Key Innovation: A brand new technique called the shift-window denoise process model, enabling auto-regressive generation for diffusion and consistency models in real-time. Special thanks to project leader Ruili Feng and the entire Matrix team for their dedication and hard work over the year-long project.

Hongyang Zhang

178,322 просмотров • 1 год назад

“real time” “30 fps” “1080p” okay (refer to keystrokes on top left for latency)
0:28

“real time” “30 fps” “1080p” okay (refer to keystrokes on top left for latency)

Iris Shaders

365,833 просмотров • 9 месяцев назад

We present MotionStream — real-time, long-duration video generation that you can interactively control just by dragging your mouse. All videos here are raw, real-time screen captures without any post-processing. Model runs on a single H100 at 29 FPS and 0.4s latency.
0:42

We present MotionStream — real-time, long-duration video generation that you can interactively control just by dragging your mouse. All videos here are raw, real-time screen captures without any post-processing. Model runs on a single H100 at 29 FPS and 0.4s latency.

Xun Huang

98,321 просмотров • 7 месяцев назад

Another beast from Alibaba. Live Avatar: real-time, infinite-length streaming avatar generator; - 20 FPS, based on a 14B WanS2V trained with DMD. - locks identity; They plan to: add TTS, low VRAM, 3 steps, SVD quantz...
1:10

Another beast from Alibaba. Live Avatar: real-time, infinite-length streaming avatar generator; - 20 FPS, based on a 14B WanS2V trained with DMD. - locks identity; They plan to: add TTS, low VRAM, 3 steps, SVD quantz...

Wildminder

21,313 просмотров • 6 месяцев назад

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 просмотров • 1 год назад

Live Avatar Streaming Real-time Audio-Driven Avatar Generation with Infinite Length
18:07

Live Avatar Streaming Real-time Audio-Driven Avatar Generation with Infinite Length

AK

22,989 просмотров • 6 месяцев назад

Introducing LTX-2: the most complete open-source AI creative engine. - Synchronized audio and video generation - Native 4K fidelity, up to 50 fps and 10 s+ sequences - API-first design for seamless integration into creative pipelines - Runs efficiently on consumer GPUs - Fully open and accessible, with weights releasing later this year
0:25

Introducing LTX-2: the most complete open-source AI creative engine. - Synchronized audio and video generation - Native 4K fidelity, up to 50 fps and 10 s+ sequences - API-first design for seamless integration into creative pipelines - Runs efficiently on consumer GPUs - Fully open and accessible, with weights releasing later this year

LTX

1,700,284 просмотров • 8 месяцев назад

Clavicular realized this generation is cooked after confronting 2 Ø/F models for ruining their life 😭
0:33

Clavicular realized this generation is cooked after confronting 2 Ø/F models for ruining their life 😭

yoxic

535,845 просмотров • 6 месяцев назад

Introducing Sparrow-1, a conversational-flow control model built for real-time conversational video that sets the benchmark for human-level conversational timing and understanding in real-time AI.
3:32

Introducing Sparrow-1, a conversational-flow control model built for real-time conversational video that sets the benchmark for human-level conversational timing and understanding in real-time AI.

Tavus

47,879 просмотров • 5 месяцев назад

⚡️ Massive Update Just Dropped: Phase 2.0 for Kling AI! 🎥 KLING 2.0 Master for video generation, 🖼️ KOLORS 2.0 for image generation, 🎮 Multi-Elements Editor, 🎨 Image Editing & Restyle... Kling AI 2.0 is all about empowering creators to bring meaningful stories to life — with powerful tools and greater creative freedom! From Vision to Screen — Made Possible by Creators Like YOU Latest updates available NOW at
1:56

⚡️ Massive Update Just Dropped: Phase 2.0 for Kling AI! 🎥 KLING 2.0 Master for video generation, 🖼️ KOLORS 2.0 for image generation, 🎮 Multi-Elements Editor, 🎨 Image Editing & Restyle... Kling AI 2.0 is all about empowering creators to bring meaningful stories to life — with powerful tools and greater creative freedom! From Vision to Screen — Made Possible by Creators Like YOU Latest updates available NOW at

Kling AI

6,957,659 просмотров • 1 год назад

We have released an update for both text to video and image to video generation with Gen-2, bringing major improvements to both the fidelity and consistency of video results. Try it now at
1:15

We have released an update for both text to video and image to video generation with Gen-2, bringing major improvements to both the fidelity and consistency of video results. Try it now at

Runway

601,381 просмотров • 2 лет назад

🌎Kling O1 is now on Dzine! We call it the Nano Banana for video generation — Edit any video as you want Repost and try it for FREE today at Dzine
0:22

🌎Kling O1 is now on Dzine! We call it the Nano Banana for video generation — Edit any video as you want Repost and try it for FREE today at Dzine

Dzine

31,390 просмотров • 6 месяцев назад

🔘 Real-time capabilities Genie 3 is our first world model to allow live interaction, while also improving consistency and realism compared to Genie 2. It can generate dynamic worlds at 720p and 24 FPS, with each frame created in response to user actions.
0:59

🔘 Real-time capabilities Genie 3 is our first world model to allow live interaction, while also improving consistency and realism compared to Genie 2. It can generate dynamic worlds at 720p and 24 FPS, with each frame created in response to user actions.

Google DeepMind

224,413 просмотров • 10 месяцев назад

Ovi is out on Hugging Face Twin Backbone Cross-Modal Fusion for Audio-Video Generation Ovi is a veo-3 like, video+audio generation model that simultaneously generates both video and audio content from text or text+image inputs. Video+Audio Generation: Generate synchronized video and audio content simultaneously Flexible Input: Supports text-only or text+image conditioning 5-second Videos: Generates 5-second videos at 24 FPS, area of 720×720, at various aspect ratios (9:16, 16:9, 1:1, etc)
2:08

Ovi is out on Hugging Face Twin Backbone Cross-Modal Fusion for Audio-Video Generation Ovi is a veo-3 like, video+audio generation model that simultaneously generates both video and audio content from text or text+image inputs. Video+Audio Generation: Generate synchronized video and audio content simultaneously Flexible Input: Supports text-only or text+image conditioning 5-second Videos: Generates 5-second videos at 24 FPS, area of 720×720, at various aspect ratios (9:16, 16:9, 1:1, etc)

AK

23,082 просмотров • 8 месяцев назад

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:
0:23

Introducing SDXL Turbo: A real-time text-to-image generation model. SDXL Turbo achieves state-of-the-art performance with a new distillation technology, enabling single-step image generation with unprecedented quality, reducing the required step count from 50 to just one. The code, research paper, and weights for non-commercial use are now available on our website. You can test SDXL Turbo on Stability AI’s image editing platform Clipdrop, with a beta demonstration of the real-time text-to-image generation capabilities. Learn more:

Stability AI

976,312 просмотров • 2 лет назад

AvatarPointillist. Autoregressive 4D Gaussian avatar generation. - drivable 3D from single portraits; - dynamically adjusts point density for hair/beards; uses DINOv2 and FLAME for animation. good identity preservation
0:37

AvatarPointillist. Autoregressive 4D Gaussian avatar generation. - drivable 3D from single portraits; - dynamically adjusts point density for hair/beards; uses DINOv2 and FLAME for animation. good identity preservation

Wildminder

18,178 просмотров • 2 месяцев назад