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.

We present MMaDA, first diffusion that unifies text reasoning, multimodal understanding, and image generation through Mixed Long-CoT, and unified RL - UniGRPO. 📚 Paper: 💻 Code: 📦 Model:

Ling Yang

4,932 subscribers

87,719 views • 1 year ago •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

11 Comments

Ling Yang1 year ago

We also provide an online demo: Welcome to use and provide suggestions!

FemMiraMancer1 year ago

MMaDA is revolutionizing the multimodal game! I'm eager to see how its unified architecture and RL algorithms elevate text-to-image generation. This could be a real milestone in AI innovation. 🚀

Stellan Haglund1 year ago

That looks pretty sequential, isn’t the upside with diffusion to predict all the tokens every step?

Ling Yang1 year ago

We use unified diffusion modeling for training and provide flexible sampling strategies (Semi-AR or Non-AR) for better performance-speed trade-off

Ruben Tous1 year ago

MMaDA is all you need.

Mr. John Smith1 year ago

Pls stop, we can’t keep up

FluffyCat1 year ago

CoT training data makes sense for autoregressive models, but what's the intuition for using it with a diffusion model?

Hyperstack1 year ago

Sounds like a game changer for bridging text and visuals 👏

Bali as a Colony of Jakarta1 year ago

not good prompt following.

Sam Woods1 year ago

That's a huge step towards truly unified multimodal AI

MangoMagic™️1 year ago

Curiosity piqued: does the model argue back? Asking for a friend. #AIorNot #DeepThoughtsNoDeepSleep #CoderInDisguise

Related Videos

Video understanding isn't just recognizing —it demands reasoning across thousands of frames. Meet Long-RL🚀 Highlights: 🧠 Dataset: LongVideo-Reason — 52K QAs with reasoning. ⚡ System: MR-SP - 2.1× faster RL for long videos. 📈 Scalability: Hour-long videos (3,600 frames) RL on a single node (8×A100s). 🖼️📝🎵 RL training for video, text, audio — works with VILA, Qwen series, and image/video generation models 🎨🎬 📄 Paper: 🎥 Demo: 💻 Code:

Video understanding isn't just recognizing —it demands reasoning across thousands of frames. Meet Long-RL🚀 Highlights: 🧠 Dataset: LongVideo-Reason — 52K QAs with reasoning. ⚡ System: MR-SP - 2.1× faster RL for long videos. 📈 Scalability: Hour-long videos (3,600 frames) RL on a single node (8×A100s). 🖼️📝🎵 RL training for video, text, audio — works with VILA, Qwen series, and image/video generation models 🎨🎬 📄 Paper: 🎥 Demo: 💻 Code:

Yukang Chen

31,855 views • 1 year ago

Let's remove VAEs and ViTs from video models! 🚀 Introducing 𝗣𝗶𝘅𝗲𝗹𝗨𝗠𝗠: an encoder-free unified multimodal model for image and video understanding and generation, directly in pixel space. Code and model available today! 🌐 📄

Let's remove VAEs and ViTs from video models! 🚀 Introducing 𝗣𝗶𝘅𝗲𝗹𝗨𝗠𝗠: an encoder-free unified multimodal model for image and video understanding and generation, directly in pixel space. Code and model available today! 🌐 📄

Cong Wei

69,297 views • 1 day ago

Introducing Omni, one unified model can support any-to-any multimodal modeling, including multimodal understanding, image/video generation and editing, world modeling and 3D reconstruction. All in one that adopts standard mixture-of-experts arch with only 3B activations.

Introducing Omni, one unified model can support any-to-any multimodal modeling, including multimodal understanding, image/video generation and editing, world modeling and 3D reconstruction. All in one that adopts standard mixture-of-experts arch with only 3B activations.

Ceyuan Yang

32,808 views • 5 months ago

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

AK

83,681 views • 3 years ago

(1/10) 🔥Thrilled to introduce OneDiffusion—our latest work in unified diffusion modeling! 🚀 This model bridges the gap between image synthesis and understanding, excelling in a wide range of tasks: T2I, conditional generation, image understanding, identity preservation, multiview generation, and even camera pose estimation. Learn more at: Project: arXiv: Code (on the way):

(1/10) 🔥Thrilled to introduce OneDiffusion—our latest work in unified diffusion modeling! 🚀 This model bridges the gap between image synthesis and understanding, excelling in a wide range of tasks: T2I, conditional generation, image understanding, identity preservation, multiview generation, and even camera pose estimation. Learn more at: Project: arXiv: Code (on the way):

Jiasen Lu

33,426 views • 1 year ago

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Tencent Hy

126,668 views • 8 months ago

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

Introducing ImagineArt 2.0 - The world's most realistic image generation model! 🧠 Reasoning-based prompt adherence 🎬 Cinematic photography ⚡ Concept to final in seconds 🔤 Text rendering that actually works This is what happens when an image model stops generating and starts reasoning.

ImagineArt

715,489 views • 5 months ago

[1/N] We present a plug-and-play mechanism to controllably steer inference of any diffusion/flow model towards a sharper or flatter sampling distribution, resulting in improvements across domains e.g. text-to-image (10% FID reduction), protein generation (improved designability).

[1/N] We present a plug-and-play mechanism to controllably steer inference of any diffusion/flow model towards a sharper or flatter sampling distribution, resulting in improvements across domains e.g. text-to-image (10% FID reduction), protein generation (improved designability).

Shubham Tulsiani

83,235 views • 1 year ago

🚀 Introducing Emu3.5 — a large-scale multimodal world model that natively predicts the next vision-language state. 🔥 Trained on over 10T interleaved vision-language tokens and enhanced with reinforcement learning, Emu3.5 achieves powerful multimodal reasoning and generation. ⚡ Powered by our new Discrete Diffusion Adaptation (DiDA) for 20× faster inference. 🔥 Emu3.5 outperforms Nano Banana across image generation, editing, interleaved tasks and more. 🌍 Explore Emu3.5: Github: #Emu3 #MultimodalAI #WorldModel #NextTokenPrediction

🚀 Introducing Emu3.5 — a large-scale multimodal world model that natively predicts the next vision-language state. 🔥 Trained on over 10T interleaved vision-language tokens and enhanced with reinforcement learning, Emu3.5 achieves powerful multimodal reasoning and generation. ⚡ Powered by our new Discrete Diffusion Adaptation (DiDA) for 20× faster inference. 🔥 Emu3.5 outperforms Nano Banana across image generation, editing, interleaved tasks and more. 🌍 Explore Emu3.5: Github: #Emu3 #MultimodalAI #WorldModel #NextTokenPrediction

BAAI

51,880 views • 11 months ago

🤯 OneDiffusion: A versatile, large-scale diffusion model that seamlessly supports bidirectional image synthesis and understanding across diverse tasks. ✅ Text to Image ✅ Image to Depth ✅ Image to Segmentation ✅ Image to Pose ✅ FaceID ✅ Image to Multiview How to use & more👇

🤯 OneDiffusion: A versatile, large-scale diffusion model that seamlessly supports bidirectional image synthesis and understanding across diverse tasks. ✅ Text to Image ✅ Image to Depth ✅ Image to Segmentation ✅ Image to Pose ✅ FaceID ✅ Image to Multiview How to use & more👇

Gradio

11,820 views • 1 year ago

Our Puffin-World technical paper is now available. From one image or text prompt, it generates camera-controlled views with appearance, geometry and physical grounding, then reconstructs them into an explorable 3D world—all in one unified model. Paper: Blog: Dataset: Code:

Our Puffin-World technical paper is now available. From one image or text prompt, it generates camera-controlled views with appearance, geometry and physical grounding, then reconstructs them into an explorable 3D world—all in one unified model. Paper: Blog: Dataset: Code:

Chen Change Loy

10,385 views • 29 days ago

Most AI demos stop at text prompts. I wanted to see what Qwen3.7-Plus could do beyond chat, so I tested it through the API using an Airbnb interface screenshot. The model analyzed the UI, identified key components, explained the visual hierarchy, highlighted usability strengths, suggested UX improvements, and even generated a React + Tailwind structure to recreate the interface. A few things that stood out: 👁️ Multimodal understanding from a single screenshot 🧠 Strong reasoning around UX and product design decisions 💻 Able to bridge visual inputs and code generation ⚡ OpenAI-compatible API that was easy to integrate 💰 Impressive capability-to-cost ratio Qwen3.7-Plus isn't just a chatbot. It's a multimodal agent model designed for workflows that combine vision, reasoning, coding, and automation. Try Qwen3.7-Plus with limited time 20% off:

Most AI demos stop at text prompts. I wanted to see what Qwen3.7-Plus could do beyond chat, so I tested it through the API using an Airbnb interface screenshot. The model analyzed the UI, identified key components, explained the visual hierarchy, highlighted usability strengths, suggested UX improvements, and even generated a React + Tailwind structure to recreate the interface. A few things that stood out: 👁️ Multimodal understanding from a single screenshot 🧠 Strong reasoning around UX and product design decisions 💻 Able to bridge visual inputs and code generation ⚡ OpenAI-compatible API that was easy to integrate 💰 Impressive capability-to-cost ratio Qwen3.7-Plus isn't just a chatbot. It's a multimodal agent model designed for workflows that combine vision, reasoning, coding, and automation. Try Qwen3.7-Plus with limited time 20% off:

Kuria | AI

17,044 views • 4 months ago

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.

Huan Ling

37,036 views • 1 year ago

You imagine it. We create it. 🎨 Introducing #PixelStudio, a first-of-its-kind image generator on #Pixel9 powered by an on-device diffusion model¹ running on Tensor G4 and our Imagen 3 text-to-image model in the cloud.

You imagine it. We create it. 🎨 Introducing #PixelStudio, a first-of-its-kind image generator on #Pixel9 powered by an on-device diffusion model¹ running on Tensor G4 and our Imagen 3 text-to-image model in the cloud.

Made by Google

28,283 views • 2 years ago

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

We've officially released and open-sourced HunyuanImage 2.1, our latest text-to-image model. The new model delivers on our commitment to balancing performance and quality. With native 2K image generation, HunyuanImage 2.1 is an advanced open-source text-to-image model.🎨 ✨ New in 2.1: 🔹Advanced Semantics: Supports ultra-long and complex prompts of up to 1000 tokens, and precisely controls the generation of multiple subjects in a single image. 🔹Precise Chinese and English Text Rendering with seamless image–text integration: The model naturally integrates text into images, making it suitable for a wide range of applications such as product covers, illustrations, and poster design to meet the needs of various fields. 🔹Rich Styles and High Aesthetic: Capable of generating images in various styles—including photorealistic portraits, comics, and vinyl figures—it delivers outstanding visual appeal and artistic quality. 🔹High-Quality Generation: Efficiently produces ultra-high-definition (2K) images in the same time other models take to generate a 1K image. HunyuanImage 2.1 uses two text encoders: a multimodal large language model (MLLM) to improve the model's image and text alignment capabilities, and a multi-language character-aware encoder to improve text rendering capabilities. The model is a single- and double-stream diffusion transformer with 17B parameters. We've also open-sourced the weights of the the accelerated version with meanflow which reduces inference steps from 100 to just 8, and PromptEnhancer, the first industrial-grade rewriting model that enhances your prompts for more nuanced and expressive image generation. Now, creators turn complex ideas—like posters with slogans or multi-panel comics—into visuals faster than ever. We’re just getting started. Stay tuned for our native multimodal image generation model coming soon. 🌐Website: 🔗Github: 🤗Hugging Face: ✨Hugging Face Demo:

Tencent Hy

89,392 views • 1 year ago

The reception since we launched the first unified audio + text + image to video model in production last week has been incredible! We wanted to take a moment and highlight some firsts: - the first omnimodal modal that jointly operates over text, image, and audio. Fun fact: we shipped our first audio-to-video model Character-1 last June! - Character-3 is generating 5 minute uncut clips, and capable of temporally stable longform prediction. - Character-3 handles realistic humans, cartoons, side profiles, dynamic backgrounds, anime, animals, and more. - Character-3 sets a new standard for fast & efficient video diffusion at HD and SD resolutions, leveraging our advancements in realtime video generation. We'll have an announcement soon, but you can read more about an earlier version of this line of work in our paper here:

The reception since we launched the first unified audio + text + image to video model in production last week has been incredible! We wanted to take a moment and highlight some firsts: - the first omnimodal modal that jointly operates over text, image, and audio. Fun fact: we shipped our first audio-to-video model Character-1 last June! - Character-3 is generating 5 minute uncut clips, and capable of temporally stable longform prediction. - Character-3 handles realistic humans, cartoons, side profiles, dynamic backgrounds, anime, animals, and more. - Character-3 sets a new standard for fast & efficient video diffusion at HD and SD resolutions, leveraging our advancements in realtime video generation. We'll have an announcement soon, but you can read more about an earlier version of this line of work in our paper here:

Hedra

180,676 views • 1 year ago

Today, we are releasing Stable Video Diffusion, our first foundation model for generative AI video based on the image model, Stable Diffusion. As part of this research preview, the code, weights, and research paper are now available. Additionally, today you can sign up for our waitlist to access a new upcoming web experience featuring a Text-To-Video interface. To access the model & sign up for our waitlist, visit our website here:

Today, we are releasing Stable Video Diffusion, our first foundation model for generative AI video based on the image model, Stable Diffusion. As part of this research preview, the code, weights, and research paper are now available. Additionally, today you can sign up for our waitlist to access a new upcoming web experience featuring a Text-To-Video interface. To access the model & sign up for our waitlist, visit our website here:

Stability AI

1,024,795 views • 2 years ago