Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

🪄 Magi-1: The Autoregressive Diffusion Video Generation Model - Now available at 🥇 The first autoregressive video model with top-tier quality output 🔓 100% open-source & tech report 📊 Exceptional performance on major benchmarks ⏳ 1/5

Sand.ai

3,678 subscribers

914,791 views • 1 year ago •via X (Twitter)

Music Science & Technology Education

Anya Rossi• Live Now

Private livecam show

10 Comments

Sand.ai1 year ago

🔑 Key Features ✅ Infinite extension, enabling seamless and comprehensive storytelling across time ✅ Offers precise control over time with one-second accuracy Going live for new web users today. Sign up at ⏳ 2/5

Sand.ai1 year ago

🍾 Opening AI for all. Proud to support the open-source community. Explore our model. 💻 Github Page: 💾 Hugging Face: ⏳ 3/5

Sand.ai1 year ago

🚰 Magi thoroughly understands the laws of physics. 🏆 Magi leads the Physics-IQ Benchmark with exceptional physics understanding ⏳ 4/5

Sand.ai1 year ago

🙇 Thank you for reading. Come and join the @magiai_hq community. 👾 ⏳ 5/5

AshutoshShrivastava1 year ago

Congratulations on the launch and great move open-sourcing it. 👏👏

Alex Volkov (Thursd/AI)1 year ago

Yoo this is sick! Will mention on @thursdai_pod this week! Welcome to send someone to tell us about this 🔥

Gil1 year ago

amazing work, couldn't find the MAGI-1-4.5B. will it be released?

chris徒.eth1 year ago

Amazing！

Robin Delta1 year ago

this is seriously next level tech wow

el.cine1 year ago

looks unreal!

Related Videos

MotionStreamer announced on Hugging Face Streaming Motion Generation via Diffusion-based Autoregressive Model in Causal Latent Space

MotionStreamer announced on Hugging Face Streaming Motion Generation via Diffusion-based Autoregressive Model in Causal Latent Space

AK

43,422 views • 1 year ago

Epona Autoregressive Diffusion World Model for Autonomous Driving

Epona Autoregressive Diffusion World Model for Autonomous Driving

AK

18,839 views • 1 year ago

Now autoregressive video models can remember. 🧠 Introducing MosaiChunk: persistent visual memory for long-horizon autoregressive video generation🧩. Instead of remembering everything, it remembers the right things. ✍️ 1/6

Now autoregressive video models can remember. 🧠 Introducing MosaiChunk: persistent visual memory for long-horizon autoregressive video generation🧩. Instead of remembering everything, it remembers the right things. ✍️ 1/6

Haven Feng ✈️ CoRL

32,548 views • 1 month ago

Video diffusion models generate high-quality videos but are too slow for interactive applications. We MIT CSAIL Adobe Research introduce CausVid, a fast autoregressive video diffusion model that starts playing the moment you hit "Generate"! A thread 🧵

Video diffusion models generate high-quality videos but are too slow for interactive applications. We MIT CSAIL Adobe Research introduce CausVid, a fast autoregressive video diffusion model that starts playing the moment you hit "Generate"! A thread 🧵

Tianwei Yin

84,074 views • 1 year ago

🔥1-min Interactive Video Generation with Multimodal Control🔥 Towards *long-context world model*, #LongVie is an end-to-end autoregressive framework for controllable ultra-long video generation - Page: - Paper: . Thanks AK

🔥1-min Interactive Video Generation with Multimodal Control🔥 Towards long-context world model, #LongVie is an end-to-end autoregressive framework for controllable ultra-long video generation - Page: - Paper: . Thanks AK

Ziwei Liu

14,213 views • 1 year ago

The first truly open-source audio-video model. LTX-2 is a DiT-based foundation model with all core video generation capabilities in one unified model. Designed to run locally on consumer GPUs. - text-to-video - image-to-video - and video-to-video modes 100% open-source.

The first truly open-source audio-video model. LTX-2 is a DiT-based foundation model with all core video generation capabilities in one unified model. Designed to run locally on consumer GPUs. - text-to-video - image-to-video - and video-to-video modes 100% open-source.

Akshay 🚀

66,268 views • 10 months ago

Can we use video diffusion to generate 3D scenes? 𝐖𝐨𝐫𝐥𝐝𝐄𝐱𝐩𝐥𝐨𝐫𝐞𝐫 (#SIGGRAPHAsia25) creates fully-navigable scenes via autoregressive video generation. Text input -> 3DGS scene output & interactive rendering! 🌍 📽️

Can we use video diffusion to generate 3D scenes? 𝐖𝐨𝐫𝐥𝐝𝐄𝐱𝐩𝐥𝐨𝐫𝐞𝐫 (#SIGGRAPHAsia25) creates fully-navigable scenes via autoregressive video generation. Text input -> 3DGS scene output & interactive rendering! 🌍 📽️

Matthias Niessner

30,942 views • 1 year ago

Researchers at Meta recently shared MAGNeT, a single non-autoregressive transformer model for text-to-music & text-to-sound generation capable of generating audio on-par with the quality of SOTA models — at 7x the speed. MAGNeT is open source as part of AudioCraft. Hear audio samples and get more details on this work ➡️

Researchers at Meta recently shared MAGNeT, a single non-autoregressive transformer model for text-to-music & text-to-sound generation capable of generating audio on-par with the quality of SOTA models — at 7x the speed. MAGNeT is open source as part of AudioCraft. Hear audio samples and get more details on this work ➡️

AI at Meta

122,001 views • 2 years ago

Rolling Forcing Autoregressive Long Video Diffusion in Real Time

Rolling Forcing Autoregressive Long Video Diffusion in Real Time

AK

30,507 views • 1 year ago

🚀 We’re excited to announce LingBot-VA, a new state-of-the-art robot policy model from Robbyant ! LingBot-VA is built on a causal, autoregressive video-action world model for generalist robot control. Highlights: (1) First unified autoregressive video-action world model for robot control (2) Low-latency inference with a new asynchronous execution pipeline (3) SOTA on RoboTwin (92.9%, firstever > 90%) and LIBERO (98.5%) (4) +20% over π0.5 on challenging real-world long-horizon & high-precision tasks

🚀 We’re excited to announce LingBot-VA, a new state-of-the-art robot policy model from Robbyant ! LingBot-VA is built on a causal, autoregressive video-action world model for generalist robot control. Highlights: (1) First unified autoregressive video-action world model for robot control (2) Low-latency inference with a new asynchronous execution pipeline (3) SOTA on RoboTwin (92.9%, firstever > 90%) and LIBERO (98.5%) (4) +20% over π0.5 on challenging real-world long-horizon & high-precision tasks

Yinghao Xu

47,815 views • 9 months ago

60-second AI video generation just got unlocked! LTXV is the first model to generate native long-form video, with controllability that beats every open source model. - 8× longer than typical gen video - 10–100× faster & cheaper - Runs even on consumer GPUs - Pose, depth & control LoRAs supported Available now. Go try it >>

60-second AI video generation just got unlocked! LTXV is the first model to generate native long-form video, with controllability that beats every open source model. - 8× longer than typical gen video - 10–100× faster & cheaper - Runs even on consumer GPUs - Pose, depth & control LoRAs supported Available now. Go try it >>

LTX

545,455 views • 1 year ago

SongBloom Coherent Song Generation via Interleaved Autoregressive Sketching and Diffusion Refinement

SongBloom Coherent Song Generation via Interleaved Autoregressive Sketching and Diffusion Refinement

AK

28,118 views • 1 year ago

MinerU-Diffusion A 2.5B diffusion-based OCR model that replaces slow autoregressive decoding with parallel block-wise diffusion, achieving up to 3.2x faster inference while improving robustness on complex documents with tables, formulas, and layouts.

MinerU-Diffusion A 2.5B diffusion-based OCR model that replaces slow autoregressive decoding with parallel block-wise diffusion, achieving up to 3.2x faster inference while improving robustness on complex documents with tables, formulas, and layouts.

DailyPapers

15,304 views • 7 months ago

We release a technical blog on training PocketTTS, our 100M parameter on-device TTS, through drifting, a recent one-step generative objective from Deng et al. Less than 1% WER, high quality and voice cloning. To the best of our knowledge, it's the first speech model, and the first autoregressive model, trained this way.

We release a technical blog on training PocketTTS, our 100M parameter on-device TTS, through drifting, a recent one-step generative objective from Deng et al. Less than 1% WER, high quality and voice cloning. To the best of our knowledge, it's the first speech model, and the first autoregressive model, trained this way.

kyutai

24,596 views • 1 month ago

LTX 2.3 is out. Another big milestone for open-source video generation. Lightricks keeps pushing the frontier with faster, higher-quality video models the community can actually build with. Open video models are accelerating fast. Model:

LTX 2.3 is out. Another big milestone for open-source video generation. Lightricks keeps pushing the frontier with faster, higher-quality video models the community can actually build with. Open video models are accelerating fast. Model:

Hugging Models

35,935 views • 8 months ago

Our research team just released Flex-Forcing: a video generation method that lets a single model switch between generation methods at inference time. Right now there are two main approaches to video generation. Bidirectional diffusion models attend to every frame at once, holding structure well at the cost of speed. Autoregressive models generate frame by frame, so they stream fast and scale to long clips, but accumulate error and drift over time. Flex-Forcing trains a single model to do both, letting you choose from the range at inference based on your compute budget.

Our research team just released Flex-Forcing: a video generation method that lets a single model switch between generation methods at inference time. Right now there are two main approaches to video generation. Bidirectional diffusion models attend to every frame at once, holding structure well at the cost of speed. Autoregressive models generate frame by frame, so they stream fast and scale to long clips, but accumulate error and drift over time. Flex-Forcing trains a single model to do both, letting you choose from the range at inference based on your compute budget.

NVIDIA AI

31,412 views • 4 months ago

PixVerse just launched their first real-time video world model R1, powered by fal's high-performance compute. It turns AI video from static clips into a real-time, interactive world. 🔄 Continuous generation stream 🧠 Autoregressive memory system ⚡ Ultra-fast response engine

PixVerse just launched their first real-time video world model R1, powered by fal's high-performance compute. It turns AI video from static clips into a real-time, interactive world. 🔄 Continuous generation stream 🧠 Autoregressive memory system ⚡ Ultra-fast response engine

fal

19,243 views • 9 months ago

Looking Ahead Modeling Future Evolution in Autoregressive Video Generation paper:

Looking Ahead Modeling Future Evolution in Autoregressive Video Generation paper:

AK

20,301 views • 21 days ago

Bilibili drops Index-AniSora on Hugging Face 👀 "The Ultimate Open-Source Anime Video Generation Model"

Bilibili drops Index-AniSora on Hugging Face 👀 "The Ultimate Open-Source Anime Video Generation Model"

Victor M

35,321 views • 1 year ago

We're moving beyond autoregressive LLMs! Autoregressive LLMs generate text word-by-word, which can be slow and affect quality, while diffusion models refine noise step-by-step, allowing for faster iterations and error correction. Here's Gemini Diffusion running at 857 tokens/s:

We're moving beyond autoregressive LLMs! Autoregressive LLMs generate text word-by-word, which can be slow and affect quality, while diffusion models refine noise step-by-step, allowing for faster iterations and error correction. Here's Gemini Diffusion running at 857 tokens/s:

Akshay 🚀

34,607 views • 1 year ago