Loading video...

Video Failed to Load

Go Home

Fast-dLLM accelerates multi-modal diffusion VLM LLaDA-V 10 times! 🚀

xieenze_jr's profile picture

Enze Xie

1,618 subscribers

11,314 views • 1 year 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

"An hour of planning can save you 10 hours of doing." ✨📝 Planned Diffusion 📝 ✨ makes a plan before parallel dLLM generation. Planned Diffusion runs 1.2-1.8× faster than autoregressive and an order of magnitude faster than diffusion, while staying within 0.9–5% AR quality.
0:21

"An hour of planning can save you 10 hours of doing." ✨📝 Planned Diffusion 📝 ✨ makes a plan before parallel dLLM generation. Planned Diffusion runs 1.2-1.8× faster than autoregressive and an order of magnitude faster than diffusion, while staying within 0.9–5% AR quality.

Daniel Israel

38,804 views • 10 months ago

🆕 Introducing JAT, the first open-source multi-modal, multi-task multi-domain agent! 🤖 A step toward open generalist agents! 🚀 📰 Blog:
0:22

🆕 Introducing JAT, the first open-source multi-modal, multi-task multi-domain agent! 🤖 A step toward open generalist agents! 🚀 📰 Blog:

Quentin Gallouédec

73,212 views • 2 years ago

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin
0:40

Speed and quality can finally coexist in diffusion-based language generation. Introducing DiDi-Instruct, a Discrete Diffusion Divergence Instruct method that distills a pre-trained discrete diffusion language model (dLLM) into a few-step student for ultra-fast generation. Built on integral KL-divergence minimization, DiDi-Instruct achieves up to 64× faster decoding, surpasses both its teacher and GPT-2, and cuts training time by 20×. Ultra-Fast Language Generation via Discrete Diffusion Divergence Instruct Paper: Code: Project: Our report: 📬 #PapersAccepted by Jiqizhixin

机器之心 JIQIZHIXIN

18,196 views • 10 months ago

Diffusion models generate high-quality images but require hundreds of forward passes. MIT CSAIL and Adobe Research introduce Distribution Matching Distillation (DMD), a distillation approach that converts costly multi-step diffusion models into fast one-step generators. A thread 🧵
0:24

Diffusion models generate high-quality images but require hundreds of forward passes. MIT CSAIL and Adobe Research introduce Distribution Matching Distillation (DMD), a distillation approach that converts costly multi-step diffusion models into fast one-step generators. A thread 🧵

MIT CSAIL

34,383 views • 2 years ago

🚀Meet Gemini Diffusion, our first diffusion-based and super fast language model, just announced at Google I/O!🚀 Very excited to be able to share what I've been working on for the past little while with our amazing small team Google DeepMind.
0:32

🚀Meet Gemini Diffusion, our first diffusion-based and super fast language model, just announced at Google I/O!🚀 Very excited to be able to share what I've been working on for the past little while with our amazing small team Google DeepMind.

Ivana Balazevic

43,562 views • 1 year ago

🚀High-Res 3D Text-to-Texture🚀 🎨#MVPaint delivers high-res 3D textures with minimal UV wrapping dependency and high versatility using synchronized multi-view diffusion 📑 Paper: 🏠 Project Page: 🧑‍💻 Code:
0:39

🚀High-Res 3D Text-to-Texture🚀 🎨#MVPaint delivers high-res 3D textures with minimal UV wrapping dependency and high versatility using synchronized multi-view diffusion 📑 Paper: 🏠 Project Page: 🧑‍💻 Code:

Ziwei Liu

28,179 views • 1 year ago

NOW you can BUILD REAL Multi-Modal Agents with our API! 🚀 Excited to announce our API is now publicly available Memories.ai 😃 Memories.ai is building the visual memory layer for AI, it enables AI to see and remember like humans do. 2025 is the year of agents, 2026 is the year of multi-modal agents! 🌟
1:03

NOW you can BUILD REAL Multi-Modal Agents with our API! 🚀 Excited to announce our API is now publicly available Memories.ai 😃 Memories.ai is building the visual memory layer for AI, it enables AI to see and remember like humans do. 2025 is the year of agents, 2026 is the year of multi-modal agents! 🌟

Shawn Shen

530,644 views • 1 year ago

Very, very fast voice bots. Llama 3.1 running on Groq Inc. 🚀 500ms voice-to-voice response times
1:52

Very, very fast voice bots. Llama 3.1 running on Groq Inc. 🚀 500ms voice-to-voice response times

kwindla

386,556 views • 2 years ago

🚀 Can blockchain oracles be fast, flexible and secure? Chainsight says yes - with zkTLS and multi-source feeds. Founder Shumpei Koike (Shumpei) explains how: 🎥
6:40

🚀 Can blockchain oracles be fast, flexible and secure? Chainsight says yes - with zkTLS and multi-source feeds. Founder Shumpei Koike (Shumpei) explains how: 🎥

DFINITY Foundation

20,182 views • 1 year ago

A tongue twister. Michael Saylor is silently stacking $KAS. Say that 10 times fast.
2:12

A tongue twister. Michael Saylor is silently stacking $KAS. Say that 10 times fast.

Tyler Bea

14,822 views • 1 year ago

Cats' reaction times are about 10-15 times faster than humans, as fast as 20 milliseconds (1/50th of a second).
0:30

Cats' reaction times are about 10-15 times faster than humans, as fast as 20 milliseconds (1/50th of a second).

Wonder of Science

340,544 views • 1 year ago

Taking a Waymo on the highway at night for the first time. As it accelerates, I realize that while I’ve been in self-driving cars many times, driving takes on a whole new meaning when it’s going fast.
0:20

Taking a Waymo on the highway at night for the first time. As it accelerates, I realize that while I’ve been in self-driving cars many times, driving takes on a whole new meaning when it’s going fast.

John Loeber 🎢

416,528 views • 9 months ago

You don’t know what God has in store for you. He’s been carrying you. He’s been sustaining you, but there are also times where God launches you forward, where He accelerates your destiny. Tune into Joel’s message today! "Fast Forward":
0:45

You don’t know what God has in store for you. He’s been carrying you. He’s been sustaining you, but there are also times where God launches you forward, where He accelerates your destiny. Tune into Joel’s message today! "Fast Forward":

Joel Osteen

95,810 views • 2 years ago

Big React Aria release! 🚀 🥳 Multi Select! 🎬 ComboBox onAction prop, e.g. "create" item ➡️ Disclosure animation ✅ New SelectionIndicator component – animations in Tabs, ToggleButtonGroup, etc. 📱 Improved modal scrolling behavior in iOS 26 📆 Calendar selectionAlignment prop
0:30

Big React Aria release! 🚀 🥳 Multi Select! 🎬 ComboBox onAction prop, e.g. "create" item ➡️ Disclosure animation ✅ New SelectionIndicator component – animations in Tabs, ToggleButtonGroup, etc. 📱 Improved modal scrolling behavior in iOS 26 📆 Calendar selectionAlignment prop

Devon Govett

51,386 views • 11 months ago

NY, we're 10 days Away⏳🚀 Tickets are going fast, make sure you get yours NOW!
0:48

NY, we're 10 days Away⏳🚀 Tickets are going fast, make sure you get yours NOW!

Baky Popile

10,487 views • 4 months ago

Exxxxciting Excalibur updates from the exceptional Mod Stu (say that 10 times fast!)...comin' your way soon!🗡️
0:52

Exxxxciting Excalibur updates from the exceptional Mod Stu (say that 10 times fast!)...comin' your way soon!🗡️

RuneScape

56,025 views • 2 years ago

Introducing MCP for arXiv Let your research agents stand on the shoulders of giants Fast multi-turn retrieval, keyword search, and embedding search tools across millions of arXiv papers 🚀
0:39

Introducing MCP for arXiv Let your research agents stand on the shoulders of giants Fast multi-turn retrieval, keyword search, and embedding search tools across millions of arXiv papers 🚀

alphaXiv

277,067 views • 6 months ago

You don’t know what God has in store for you. He’s been carrying you. He’s been sustaining you, but there are also times when God catapults you ahead, when He accelerates your destiny. Check out this message from Joel! "Fast Forward":
0:45

You don’t know what God has in store for you. He’s been carrying you. He’s been sustaining you, but there are also times when God catapults you ahead, when He accelerates your destiny. Check out this message from Joel! "Fast Forward":

Joel Osteen

20,109 views • 4 months ago

MAX-V + LB LIFT Multi-Directional Plyos (BIGs) Arc Run Race (Skill/Mids) Serpentine Run Chase Fly 10 (20yd build) (BIGs) (10yd build) LB Lift
1:20

MAX-V + LB LIFT Multi-Directional Plyos (BIGs) Arc Run Race (Skill/Mids) Serpentine Run Chase Fly 10 (20yd build) (BIGs) (10yd build) LB Lift

Nicholas Giblin

87,246 views • 1 year ago