Загрузка видео...

Не удалось загрузить видео

На главную

📦 Fully open-source under Apache 2.0 — the FULL stack: ✅ Base model (256p) ✅ Distilled model ✅ Super-resolution model (540p & 1080p) ✅ Inference code 🤗 Models: 🎮 Demo: 💻 Code: 📄 Paper: More videos below! Go build something amazing 🚀

stefan_fee's profile picture

Pengfei Liu

5,020 subscribers

12,163 просмотров • 5 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 6

Фото профиля Pengfei Liu
Pengfei Liu5 месяцев назад

📊 Benchmarks — 2,000 pairwise human comparisons: 🥊 vs Ovi 1.1 → 80.0% win rate 💥 🥊 vs LTX 2.3 → 60.9% win rate 🔥 🎨 Visual Quality: 4.80 (highest) 📝 Text Alignment: 4.18 (highest) 🎙️ WER: 14.60% (lowest) Opensource SOTA across the board 👑

Фото профиля Pengfei Liu
Pengfei Liu5 месяцев назад

🧑 Human-centric quality that hits different: 😊 Expressive facial performance 🗣️ Natural speech-expression coordination 🕺 Realistic body motion 🎵 Accurate audio-video sync 🌍 Multilingual: Chinese (Mandarin + Cantonese), English, Japanese, Korean, German, French 📉 WER of just 14.6% — lowest among all competitors

Фото профиля Pengfei Liu
Pengfei Liu5 месяцев назад

🏎️ Why it's SO fast: ⚡ Latent-space super-res — upscale in latent space, and skip the extra VAE decode-encode round trip 🔄 Turbo VAE Decoder — lightweight retrained decoder slashes decoding overhead 🔧 MagiCompiler — full-graph compilation fuses ops across layers (~1.2x speedup) 🎯 DMD-2 distillation → only 8 denoising steps, no CFG 256p in 2s ⚡ 540p in 8s ⚡ 1080p in 38s All on a single H100 🤯

Фото профиля Pengfei Liu
Pengfei Liu5 месяцев назад

🏗️ Architecture deep dive: A single 15B-param, 40-layer Transformer processes text, video, and audio tokens through self-attention only. 🥪 "Sandwich" design: first & last 4 layers use modality-specific projections, middle 32 layers share params across all modalities. 🚫 No timestep embeddings — the model infers denoising state directly from input latents.

Фото профиля Pengfei Liu
Pengfei Liu5 месяцев назад

Seedance 2.0 is impressive. But it's closed-source! Introducing our daVinci-MagiHuman — a single-stream 15B Transformer trained from scratch that jointly generates video + audio. No cross-attention. No multi-stream branches. Just self-attention. ⚡ 5s 1080p video in 38s on a single H100 🏆 80% win rate vs Ovi 1.1 | 60.9% vs LTX 2.3 (2,000 human comparisons) 🌍 6 languages 📦 Fully open-source Speed by simplicity. By @SII_GAIR × @SandAI_HQ 📄 💻 🤗

Фото профиля Surendra Anubolu
Surendra Anubolu5 месяцев назад

Impressive! Has anyone tried this on 5090? How long on 5090 for a 5s video

Похожие видео

🚀 Hunyuan 3D 2.1 is here! The first fully open-source, production-ready PBR 3D generative model! ✅Cinema-grade visuals: PBR material synthesis brings leather, bronze, and more to life with stunning light interactions. ✅ Fully open-source: Model weights, training/inference code, data pipelines, and architecture—yours to fine-tune! ✅ Runs on consumer GPUs, empowering creators, devs, and small teams. Be the first to shape the future of 3D! Download now and build something epic. Model: Demo: Github: Hunyuan 3D Creation Engine: #Hunyuan3D #OpenSource #3DCreation
3:05

🚀 Hunyuan 3D 2.1 is here! The first fully open-source, production-ready PBR 3D generative model! ✅Cinema-grade visuals: PBR material synthesis brings leather, bronze, and more to life with stunning light interactions. ✅ Fully open-source: Model weights, training/inference code, data pipelines, and architecture—yours to fine-tune! ✅ Runs on consumer GPUs, empowering creators, devs, and small teams. Be the first to shape the future of 3D! Download now and build something epic. Model: Demo: Github: Hunyuan 3D Creation Engine: #Hunyuan3D #OpenSource #3DCreation

Tencent Hy

236,398 просмотров • 1 год назад

The new episode of #TheAgentFactory is live! We’re diving into Gemma 4, the powerhouse open model family for single GPUs. See it in action: ✅ Apache 2.0 license ✅ Native function calling ✅ 100% offline agents And, watch it build in real time →
0:22

The new episode of #TheAgentFactory is live! We’re diving into Gemma 4, the powerhouse open model family for single GPUs. See it in action: ✅ Apache 2.0 license ✅ Native function calling ✅ 100% offline agents And, watch it build in real time →

Google Cloud Tech

15,442 просмотров • 5 месяцев назад

🚀 We open-sourced LongLive — interactive, real-time long-video generation. 👥Generates video in real time as users enter text prompts. ⚡️20.7 FPS on a single H100,⏱️up to 240s per clip. 🎬Fine-tunes SOTA short-video models (e.g., Wan) into long-video generators. 🌍One step closer to World Models. All code for training & inference, model weights, demo page, and videos released! Paper: Code: Model: Demo Page: Introduction Video:
3:18

🚀 We open-sourced LongLive — interactive, real-time long-video generation. 👥Generates video in real time as users enter text prompts. ⚡️20.7 FPS on a single H100,⏱️up to 240s per clip. 🎬Fine-tunes SOTA short-video models (e.g., Wan) into long-video generators. 🌍One step closer to World Models. All code for training & inference, model weights, demo page, and videos released! Paper: Code: Model: Demo Page: Introduction Video:

Yukang Chen

11,943 просмотров • 11 месяцев назад

🚀 New on ModelScope: MiniMax M2.1 is open-source! ✅ SOTA in 8+ languages (Rust, Go, Java, C++, TS, Kotlin, Obj-C, JS) ✅ Full-stack Web & mobile dev: Android/iOS, 3D visuals, vibe coding that actually ships ✅ Smarter, faster, 30% fewer tokens — with lightning mode (M2.1-lightning) for high-TPS workflows ✅ Top-tier on SWE-bench, VIBE, and custom coding/review benchmarks ✅ Works flawlessly in Cursor, Cline, Droid, BlackBox, and more It’s not just “better code” — it’s AI-native development, end to end. 🔗 Model:
0:54

🚀 New on ModelScope: MiniMax M2.1 is open-source! ✅ SOTA in 8+ languages (Rust, Go, Java, C++, TS, Kotlin, Obj-C, JS) ✅ Full-stack Web & mobile dev: Android/iOS, 3D visuals, vibe coding that actually ships ✅ Smarter, faster, 30% fewer tokens — with lightning mode (M2.1-lightning) for high-TPS workflows ✅ Top-tier on SWE-bench, VIBE, and custom coding/review benchmarks ✅ Works flawlessly in Cursor, Cline, Droid, BlackBox, and more It’s not just “better code” — it’s AI-native development, end to end. 🔗 Model:

ModelScope

16,939 просмотров • 8 месяцев назад

Spatial Intelligence just went Open Source. 🌐 Introducing InSpatio-WorldFM — the world’s first open-source real-time interactive 3D world model. Last October, World Labs's RTFM showed the vision. Today, we’re bringing that idea into the open for everyone. ✅Efficient, multi-view consistent spatial reasoning ✅Real-time interactive exploration ✅Open-source weights & code Stop watching demos. Start building now. 👇
0:54

Spatial Intelligence just went Open Source. 🌐 Introducing InSpatio-WorldFM — the world’s first open-source real-time interactive 3D world model. Last October, World Labs's RTFM showed the vision. Today, we’re bringing that idea into the open for everyone. ✅Efficient, multi-view consistent spatial reasoning ✅Real-time interactive exploration ✅Open-source weights & code Stop watching demos. Start building now. 👇

InSpatio｜影溯

37,396 просмотров • 6 месяцев назад

1/5 🚀 Thrilled to open-source OSCAR 🤖 — an action-conditioned world model for robotics, led by the visiting student in my group Zhuoyuan Wu! It generalizes across different robot embodiments with precise action controllability. All trained on a single GH200 GPU, and outperforms existing open-sourced baselines, which have larger model capacity and need more compute. Everything is public, including training data. 📄 Paper: 🌐 Project: 💻 Code: 🤗 Robot data: 🤗 Human data: 🤗 Weights: #Robotics #WorldModels #AI #OpenSource
0:31

1/5 🚀 Thrilled to open-source OSCAR 🤖 — an action-conditioned world model for robotics, led by the visiting student in my group Zhuoyuan Wu! It generalizes across different robot embodiments with precise action controllability. All trained on a single GH200 GPU, and outperforms existing open-sourced baselines, which have larger model capacity and need more compute. Everything is public, including training data. 📄 Paper: 🌐 Project: 💻 Code: 🤗 Robot data: 🤗 Human data: 🤗 Weights: #Robotics #WorldModels #AI #OpenSource

Jun Gao

104,967 просмотров • 3 месяцев назад

🚀 Introducing KumoRFM — the world’s first Relational Foundation Model purpose-built for enterprise prediction tasks! KumoRFM reasons over complex relational data to deliver instant, accurate, in-context predictions — no task-specific model training required. A true game-changer for solving key business problems like: ✅ Product recommendations ✅ Fraud detection ✅ Customer retention 🔗 Explore KumoRFM: 📄 Read the paper: 💡 Learn more: #AI #EnterpriseAI #RelationalAI #FoundationModels #MachineLearning #KumoRFM #PredictiveAI4o
0:58

🚀 Introducing KumoRFM — the world’s first Relational Foundation Model purpose-built for enterprise prediction tasks! KumoRFM reasons over complex relational data to deliver instant, accurate, in-context predictions — no task-specific model training required. A true game-changer for solving key business problems like: ✅ Product recommendations ✅ Fraud detection ✅ Customer retention 🔗 Explore KumoRFM: 📄 Read the paper: 💡 Learn more: #AI #EnterpriseAI #RelationalAI #FoundationModels #MachineLearning #KumoRFM #PredictiveAI4o

Jure Leskovec

63,728 просмотров • 1 год назад

Introducing Open TTS Tracker! 🗣️ *sound on* A one-stop shop to track all open access/ source TTS models! Ranging from XTTS to Pheme, OpenVoice to VITS, and more... ⚡ For each model, we compile: 1. Souce-code 2. Checkpoints 3. License 4. Fine-tuning code 5. Languages supported 6. Paper 7. Demo Help us make it more complete! Let's 2024 the year of open TTS models! ❤️
1:27

Introducing Open TTS Tracker! 🗣️ *sound on* A one-stop shop to track all open access/ source TTS models! Ranging from XTTS to Pheme, OpenVoice to VITS, and more... ⚡ For each model, we compile: 1. Souce-code 2. Checkpoints 3. License 4. Fine-tuning code 5. Languages supported 6. Paper 7. Demo Help us make it more complete! Let's 2024 the year of open TTS models! ❤️

Vaibhav (VB) Srivastav

68,064 просмотров • 2 лет назад

Introducing Ollama fully Integrated into VSCode with CodeGPT Now you can download open-source models from ollama with a single click and install them as code assistants in VSCode. In this case, I download the qwen2:0.5b (Qwen) model in under a minute, and it becomes fully functional with just one click. The era of locally running code assistants has begun! 🚀
1:14

Introducing Ollama fully Integrated into VSCode with CodeGPT Now you can download open-source models from ollama with a single click and install them as code assistants in VSCode. In this case, I download the qwen2:0.5b (Qwen) model in under a minute, and it becomes fully functional with just one click. The era of locally running code assistants has begun! 🚀

Daniel San

101,567 просмотров • 2 лет назад

🚀 LTXV 0.9.5 is here! Our latest AI video model brings: ✅ Commercial licensing 🎬 Keyframe conditioning 🔍 Higher resolution & improved quality 📈 Longer sequences, fewer artifacts 🖥️ Native ComfyUI support Bringing open-source AI video creation to the next level.
0:53

🚀 LTXV 0.9.5 is here! Our latest AI video model brings: ✅ Commercial licensing 🎬 Keyframe conditioning 🔍 Higher resolution & improved quality 📈 Longer sequences, fewer artifacts 🖥️ Native ComfyUI support Bringing open-source AI video creation to the next level.

LTX.io

65,581 просмотров • 1 год назад

🚀 LTXV 0.9.5 is here! Our latest AI video model brings: ✅ Commercial licensing 🎬 Keyframe conditioning 🔍 Higher resolution & improved quality 📈 Longer sequences, fewer artifacts 🖥️ Native ComfyUI support Bringing open-source AI video creation to the next level.
0:53

🚀 LTXV 0.9.5 is here! Our latest AI video model brings: ✅ Commercial licensing 🎬 Keyframe conditioning 🔍 Higher resolution & improved quality 📈 Longer sequences, fewer artifacts 🖥️ Native ComfyUI support Bringing open-source AI video creation to the next level.

Lightricks

31,740 просмотров • 1 год назад

.Cohere is defining the future of sovereign AI. Autumn Moulder, VP of Engineering at Cohere, breaks down the requirements for a full AI sovereign stack at #NVIDIAGTC: ✅ Full-Stack Sovereignty: Deploying everything from the model to applications and reasoning traces within a single data center. ✅ The Power of Openness: Why open models like NVIDIA Nemotron are critical for data lineage and navigating evolving regulations.
1:05

.Cohere is defining the future of sovereign AI. Autumn Moulder, VP of Engineering at Cohere, breaks down the requirements for a full AI sovereign stack at #NVIDIAGTC: ✅ Full-Stack Sovereignty: Deploying everything from the model to applications and reasoning traces within a single data center. ✅ The Power of Openness: Why open models like NVIDIA Nemotron are critical for data lineage and navigating evolving regulations.

NVIDIA AI

10,965 просмотров • 5 месяцев назад

🚨 Breaking: Google will soon launch Gemini CLI, a powerful open-source AI agent built for the terminal. - Built on Gemini 2.5 Pro - 1 million token context window - Free tier with 60 requests per minute and 1,000 per day - Google Search grounding for real-time context - Script and plugin support - Non-interactive mode for automation - Support for Model Context Protocol (MCP) - Integrated with Gemini Code Assist in VS Code - Fully open-source under Apache 2.0
0:22

🚨 Breaking: Google will soon launch Gemini CLI, a powerful open-source AI agent built for the terminal. - Built on Gemini 2.5 Pro - 1 million token context window - Free tier with 60 requests per minute and 1,000 per day - Google Search grounding for real-time context - Script and plugin support - Non-interactive mode for automation - Support for Model Context Protocol (MCP) - Integrated with Gemini Code Assist in VS Code - Fully open-source under Apache 2.0

AshutoshShrivastava

217,433 просмотров • 1 год назад

Introducing Agent S2 – Our newest open-source AI agent setting new records in computer & smartphone use! We are seeing Agent S2 solve a whole new range of tasks, pushing the boundaries of AI-driven autonomy. 🔥 Why it’s special: ✅ #1 in OSWorld (34.5% accuracy at 50 steps, surpassing OpenAI’s Operator) ✅ #1 in AndroidWorld (50% accuracy, advancing state-of-the-art by +3.4%) ✅ Fully Open-Source & Free to Use 🔗 Blog: 🔗 Code: 📄 Paper coming soon! Join us & shape the future of personal agents. Start building today! 🛠️
0:29

Introducing Agent S2 – Our newest open-source AI agent setting new records in computer & smartphone use! We are seeing Agent S2 solve a whole new range of tasks, pushing the boundaries of AI-driven autonomy. 🔥 Why it’s special: ✅ #1 in OSWorld (34.5% accuracy at 50 steps, surpassing OpenAI’s Operator) ✅ #1 in AndroidWorld (50% accuracy, advancing state-of-the-art by +3.4%) ✅ Fully Open-Source & Free to Use 🔗 Blog: 🔗 Code: 📄 Paper coming soon! Join us & shape the future of personal agents. Start building today! 🛠️

Simular

221,781 просмотров • 1 год назад

LETS GOO! Parler TTS 🔥 A fully open-source, Apache 2.0 licensed Text-to-speech model focused on providing maximum controllability. Through voice prompts, you can control the pitch, speed, gender, noise levels, emotion characteristics and more! > Trained on 10K hours of permissive data. > Offers control over the generations. > Training + Inference code released. > The processed dataset and tagging scripts were released for further research. > English only for now. Next, we're scaling the training to 50K hours and even better dataset processing! Want to help us out? DMs open! 🤗
0:39

LETS GOO! Parler TTS 🔥 A fully open-source, Apache 2.0 licensed Text-to-speech model focused on providing maximum controllability. Through voice prompts, you can control the pitch, speed, gender, noise levels, emotion characteristics and more! > Trained on 10K hours of permissive data. > Offers control over the generations. > Training + Inference code released. > The processed dataset and tagging scripts were released for further research. > English only for now. Next, we're scaling the training to 50K hours and even better dataset processing! Want to help us out? DMs open! 🤗

Vaibhav (VB) Srivastav

156,386 просмотров • 2 лет назад

We are releasing Falcon Perception, an open-vocabulary referring expression segmentation model. Along with it, a 0.3B OCR model that is on par with 3-10x larger competitors. Current systems solve this with complex pipelines (separate encoders, late fusion, matching algorithms). We developed a novel simpler "bitter" approach: one early-fusion Transformer (image + text from first layer) with a shared parameter space, and let scale + training signal do the work. Please check our work ! 📄 Paper: 💻 Code: 🎮 Playground: 🤗 Blogpost:
1:10

We are releasing Falcon Perception, an open-vocabulary referring expression segmentation model. Along with it, a 0.3B OCR model that is on par with 3-10x larger competitors. Current systems solve this with complex pipelines (separate encoders, late fusion, matching algorithms). We developed a novel simpler "bitter" approach: one early-fusion Transformer (image + text from first layer) with a shared parameter space, and let scale + training signal do the work. Please check our work ! 📄 Paper: 💻 Code: 🎮 Playground: 🤗 Blogpost:

Yasser Dahou

121,687 просмотров • 5 месяцев назад