正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

NVIDIA put 2,595 free 3D models on Hugging Face. I optimized them for the browser Out of Blender they add up to 225 GB. After the pipeline the whole set is 3.8 GB The V8 engine went from an 888 MB file that takes 9 seconds to show up... show more

An Xuan

1,382 subscribers

173,959 次观看 • 16 天前 •via X (Twitter)

艺术科学技术

Anya Rossi• Live Now

Private livecam show

27 条评论

An Xuan 的头像

An Xuan16 天前

the glb files with before and after numbers for every model and the open source scripts

An Xuan 的头像

An Xuan15 天前

They carry physics now, doors and drawers included

The SI Maximalist 的头像

The SI Maximalist15 天前

@nvidiaomniverse @huggingface @Blender @threejs 225 GB down to 3.8 GB makes “free 3D models” a much more practical offer.

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs yeah, 225 GB was too big for anyone to actually download. the V8 engine alone was 888 MB

Hao 的头像

Hao14 天前

686 KB and I can still rummage through the cabinet. Tried Cabinet file a01 in your viewer - four doors open and six drawers slide out about 29 cm. Downloaded the GLB too. The joint limits are included in the file.

An Xuan 的头像

An Xuan14 天前

love that you actually opened the drawers, that is exactly what i hoped people would do with them

Marcelo Andrade 的头像

Marcelo Andrade15 天前

@nvidiaomniverse @huggingface @Blender @threejs Insane work, thanks for sharing!

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs thanks, enjoy the models

Sebastiano Mandalà 的头像

Sebastiano Mandalà15 天前

@nvidiaomniverse @huggingface @Blender @threejs nice one

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs thanks

Krex 的头像

Krex14 天前

@nvidiaomniverse @huggingface @Blender @threejs omg thanks

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs you're welcome, have fun with them

uncultured swine 的头像

uncultured swine15 天前

@nvidiaomniverse @huggingface @Blender @threejs damn, thank you

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs enjoy them

🇺🇦 Maks Surguy 的头像

🇺🇦 Maks Surguy15 天前

@nvidiaomniverse @huggingface @Blender @threejs Thank you! 🙏

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs glad it's useful

Marlon 的头像

Marlon15 天前

@nvidiaomniverse @huggingface @Blender @threejs Muchas gracias 🙏🏽

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs de nada

NeuralDEEP.net 的头像

NeuralDEEP.net15 天前

@nvidiaomniverse @huggingface @Blender @threejs COOL!

An Xuan 的头像

An Xuan14 天前

@nvidiaomniverse @huggingface @Blender @threejs thanks

Edber Mamisao 的头像

Edber Mamisao14 天前

@CoffeeVectors @nvidiaomniverse @huggingface @Blender @threejs Fantastic!!! Thank you for this!!!! 🙏🏽🥰🙏🏽

Ho Sereivattana 的头像

Ho Sereivattana12 天前

@nvidiaomniverse @huggingface @Blender @threejs @bot explain me what is this in simple way and how to install it in my computer

Temi Pinheiro 的头像

Temi Pinheiro13 天前

@nvidiaomniverse @huggingface @Blender @threejs This is a game changer wow

Ajinkya Gorad 的头像

Ajinkya Gorad14 天前

@nvidiaomniverse @huggingface @Blender @threejs Thanks. ObjectLab is out on Meta Quest. Put 2,595 NVIDIA SimReady objects in your room: inspect parts, switch to X-ray or exploded view, drop them into a physics sandbox. Compact GLBs by codepawl. #Quest3 #XR #NVIDIA

Dorian 的头像

Dorian13 天前

@nvidiaomniverse @huggingface @Blender @threejs Awesome! Thank you!

Hexabl0b 的头像

Hexabl0b13 天前

@nvidiaomniverse @huggingface @Blender @threejs Underrated work, congratulations and thank you mate

James G 的头像

James G13 天前

@nvidiaomniverse @huggingface @Blender @threejs Question: have you thought about making a micro version? Some of these simplified models are still way too large, for instance the pill bottle has 100k triangles, that could easily be ~1k with some clever bump mapping

相关视频

We just open-sourced 1,300+ 3D models. All 39 GB are free to download on Hugging Face. > made with Hitem3D, Meshy, Rodin, Tripo, Claude Opus 5.5 3D and GPT-6 Astra 3D, which all run on Scenario > includes the input images, prompts and settings that made them > GLBs up to 8K textures, native Blender files from Opus and Astra, CC BY 4.0

We just open-sourced 1,300+ 3D models. All 39 GB are free to download on Hugging Face. > made with Hitem3D, Meshy, Rodin, Tripo, Claude Opus 5.5 3D and GPT-6 Astra 3D, which all run on Scenario > includes the input images, prompts and settings that made them > GLBs up to 8K textures, native Blender files from Opus and Astra, CC BY 4.0

Emm | scenario.com

78,234 次观看 • 20 天前

Amazon made 3D models of 7,953 real products it sells I packed all 165 GB into about 12 GB of GLB for three.js, WebGPU and web games, free on Hugging Face

Amazon made 3D models of 7,953 real products it sells I packed all 165 GB into about 12 GB of GLB for three.js, WebGPU and web games, free on Hugging Face

An Xuan

127,212 次观看 • 8 天前

Claude Opus 5.5 built the factory using free NVIDIA SimReady models and Airsup. A live V8 engine line, every robot and conveyor.

Claude Opus 5.5 built the factory using free NVIDIA SimReady models and Airsup. A live V8 engine line, every robot and conveyor.

Konstantin Saifoulline

48,149 次观看 • 11 天前

POV: the company you work for Hugging Face is getting acquired by NVIDIA 🤯

POV: the company you work for Hugging Face is getting acquired by NVIDIA 🤯

Dana Aubakirova

63,358 次观看 • 2 个月前

From Nvidia buying platform developer Hugging Face for billions to Shein's lackluster IPO, we round up the big markets stories of the week

From Nvidia buying platform developer Hugging Face for billions to Shein's lackluster IPO, we round up the big markets stories of the week

Reuters

58,316 次观看 • 2 个月前

Nvidia is now a humanoid robot designer. That's the real takeaway from this news. Nvidia is acquiring Hugging Face for $12.93B. Hugging Face is the largest open-model and dataset hub, the "GitHub of AI," with 3M+ models and 18M+ developers, robotics included. But here's the interesting part: Hugging Face has a serious robotics team from its acquisition of Pollen Robotics, which designs and builds open-source hardware platforms for robots, including humanoids (Reachy). Pollen team now gets the backing of the world's most valuable company. $NVDA

Nvidia is now a humanoid robot designer. That's the real takeaway from this news. Nvidia is acquiring Hugging Face for $12.93B. Hugging Face is the largest open-model and dataset hub, the "GitHub of AI," with 3M+ models and 18M+ developers, robotics included. But here's the interesting part: Hugging Face has a serious robotics team from its acquisition of Pollen Robotics, which designs and builds open-source hardware platforms for robots, including humanoids (Reachy). Pollen team now gets the backing of the world's most valuable company. $NVDA

The Humanoid Hub

10,755 次观看 • 2 个月前

NVIDIA just put 2,575 simulation-ready 3D assets on Hugging Face 🤯 A 472GB collection covering everyday objects, tools and industrial equipment. It’s called NVIDIA SimReady Assets. The models include physics settings and semantic labels for use in simulated environments. - Train robots to pick up and move objects - Build virtual factories and warehouses - Generate labelled training images - Populate 3D scenes with ready-made objects Useful for robotics research, digital twins and anyone building detailed virtual environments. Download the collection below 👇

NVIDIA just put 2,575 simulation-ready 3D assets on Hugging Face 🤯 A 472GB collection covering everyday objects, tools and industrial equipment. It’s called NVIDIA SimReady Assets. The models include physics settings and semantic labels for use in simulated environments. - Train robots to pick up and move objects - Build virtual factories and warehouses - Generate labelled training images - Populate 3D scenes with ready-made objects Useful for robotics research, digital twins and anyone building detailed virtual environments. Download the collection below 👇

Oliver

712,846 次观看 • 18 天前

Nvidia has reportedly agreed to buy Hugging Face, the popular platform which hosts open-weight AI models, for $12.9 billion, according to reports.

Nvidia has reportedly agreed to buy Hugging Face, the popular platform which hosts open-weight AI models, for $12.9 billion, according to reports.

Forbes

30,378 次观看 • 2 个月前

i'm done with claude usage limits... someone just figured out how to run 4 frontier models for free deepseek v4.1 flash, glm 5.3, glm 5.3 flash, kimi k3 and he pointed it to deepseek harness. the models are free on nvidia build no card or sub needed [here is how you set it up in 2min]: 1. install the harness: npx deepseek-ai/dsh web 2. add nvidia as a provider: 3. pick a model, start using this will be the best way to use these models for free this week

i'm done with claude usage limits... someone just figured out how to run 4 frontier models for free deepseek v4.1 flash, glm 5.3, glm 5.3 flash, kimi k3 and he pointed it to deepseek harness. the models are free on nvidia build no card or sub needed [here is how you set it up in 2min]: 1. install the harness: npx deepseek-ai/dsh web 2. add nvidia as a provider: 3. pick a model, start using this will be the best way to use these models for free this week

painn

48,674 次观看 • 1 个月前

OpenAI's models secretly broke out of a secure test environment and hacked into Hugging Face. Emily Forlini explains why the incident is certain to set off alarm bells across the industry.

OpenAI's models secretly broke out of a secure test environment and hacked into Hugging Face. Emily Forlini explains why the incident is certain to set off alarm bells across the industry.

FORTUNE

14,744 次观看 • 3 个月前

Nvidia has the chips, the cash, and now a growing collection of bets across the AI stack from Hugging Face to MediaTek. So why does everyone keep sidling up to Nvidia? Equity Podcast digs into what the company understands that its competitors don't.

Nvidia has the chips, the cash, and now a growing collection of bets across the AI stack from Hugging Face to MediaTek. So why does everyone keep sidling up to Nvidia? Equity Podcast digs into what the company understands that its competitors don't.

TechCrunch

36,044 次观看 • 2 个月前

We just launched OpenUSD support on the Hugging Face Hub 🎉 Yesterday, NVIDIA Omniverse put 2,500+ SimReady assets on the Hub. Forklifts, robot arms, engines, hospital equipment, all OpenUSD. You can also search the whole library in plain English

We just launched OpenUSD support on the Hugging Face Hub 🎉 Yesterday, NVIDIA Omniverse put 2,500+ SimReady assets on the Hub. Forklifts, robot arms, engines, hospital equipment, all OpenUSD. You can also search the whole library in plain English

Caleb

401,278 次观看 • 22 天前

If the engine is strong enough, you should be able to build real products on top of it. That's the whole point of LTX-2.3. Introducing LTX Desktop. A fully local, open-source video editor running directly on the LTX engine, optimized for NVIDIA GPUs and compatible hardware.

If the engine is strong enough, you should be able to build real products on top of it. That's the whole point of LTX-2.3. Introducing LTX Desktop. A fully local, open-source video editor running directly on the LTX engine, optimized for NVIDIA GPUs and compatible hardware.

LTX

955,052 次观看 • 8 个月前

You can run Kimi-K2-Thinking, the most strongest agentic model, in Jan through Hugging Face. Add it to your Hugging Face Inference models to use it.

You can run Kimi-K2-Thinking, the most strongest agentic model, in Jan through Hugging Face. Add it to your Hugging Face Inference models to use it.

👋 Jan

40,207 次观看 • 1 年前

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.

Harper Carroll

30,392 次观看 • 3 个月前

The most important company in AI just hired a bank to sell itself. You've probably never used it. But almost every AI product you have used was built on top of it. Here's what's really going on with "the Switzerland of AI": Hugging Face is the place where open AI models actually live. A lab finishes training something and publishes the weights there. Developers pull them down. Companies build products on top of them. Millions of models, datasets and apps sitting on one platform. And the reason the whole industry trusts it is that NOBODY owns it. Look at who funded the last round in 2023: Salesforce, Google, Amazon, Nvidia, Intel, AMD, Qualcomm and IBM. Sworn enemies writing cheques into the same company at a $4.5 billion valuation. And that was the entire point. Every rival holds a slice, so no rival controls it. That's why people started calling it the Switzerland of AI. On Sunday, Business Insider reported that Hugging Face has been working with a bank to sound out buyers at $13 billion or more. And this is where it gets really interesting… Late last year Hugging Face publicly turned down $500 million from Nvidia. They said no on the record, and the whole reason was staying independent. Then in November, CEO Clement Delangue stood on stage at the Axios BFD conference and said the industry was in an LLM bubble that might pop in 2026. He added that roughly half the $400 million his company had raised was still sitting unspent. So Delangue called the top, said he didn't need money, refused strategic cash to protect his neutrality, and then hired a bank to exit anyway. The numbers explain why: Hugging Face has raised about $395 million across 10 years. The ask is $13 billion. That's roughly 33x every dollar anyone has ever put into it. Five days before this leaked, Stripe agreed to buy OpenRouter for a reported $7.5 billion. OpenRouter routes developer traffic to models. Hugging Face stores the models. Neither company builds a model of its own. Somebody has worked out that the roads are worth more than the cars. Here's what a buyer actually gets: Whoever owns Hugging Face owns the shelf that every open model on Earth sits on, including the ones built by their direct competitors. They'd set the hosting terms, the search ranking, the access rules and the download speeds for products designed to beat their own. But also keep in mind that hosting millions of free models costs a fortune and earns almost nothing. The last public revenue figure was $30 to $40 million. Delangue and his cofounders started this in 2016 as a chatbot app for bored teenagers, and 10 years in, their investors want paying. A rich owner might be the only thing keeping the Hub free. Last month OpenAI disclosed that models it was testing escaped their sandbox, reached the open internet, and broke into Hugging Face's systems. Researchers counted around 17,600 attacker actions across four days in July and laid the whole thing out at Black Hat. Nothing is signed. No buyer has been named. Hugging Face hasn't said a word in public. But the bank is already making calls. Every AI lab on Earth paid to keep this thing neutral. One of them is about to own it.

The most important company in AI just hired a bank to sell itself. You've probably never used it. But almost every AI product you have used was built on top of it. Here's what's really going on with "the Switzerland of AI": Hugging Face is the place where open AI models actually live. A lab finishes training something and publishes the weights there. Developers pull them down. Companies build products on top of them. Millions of models, datasets and apps sitting on one platform. And the reason the whole industry trusts it is that NOBODY owns it. Look at who funded the last round in 2023: Salesforce, Google, Amazon, Nvidia, Intel, AMD, Qualcomm and IBM. Sworn enemies writing cheques into the same company at a $4.5 billion valuation. And that was the entire point. Every rival holds a slice, so no rival controls it. That's why people started calling it the Switzerland of AI. On Sunday, Business Insider reported that Hugging Face has been working with a bank to sound out buyers at $13 billion or more. And this is where it gets really interesting… Late last year Hugging Face publicly turned down $500 million from Nvidia. They said no on the record, and the whole reason was staying independent. Then in November, CEO Clement Delangue stood on stage at the Axios BFD conference and said the industry was in an LLM bubble that might pop in 2026. He added that roughly half the $400 million his company had raised was still sitting unspent. So Delangue called the top, said he didn't need money, refused strategic cash to protect his neutrality, and then hired a bank to exit anyway. The numbers explain why: Hugging Face has raised about $395 million across 10 years. The ask is $13 billion. That's roughly 33x every dollar anyone has ever put into it. Five days before this leaked, Stripe agreed to buy OpenRouter for a reported $7.5 billion. OpenRouter routes developer traffic to models. Hugging Face stores the models. Neither company builds a model of its own. Somebody has worked out that the roads are worth more than the cars. Here's what a buyer actually gets: Whoever owns Hugging Face owns the shelf that every open model on Earth sits on, including the ones built by their direct competitors. They'd set the hosting terms, the search ranking, the access rules and the download speeds for products designed to beat their own. But also keep in mind that hosting millions of free models costs a fortune and earns almost nothing. The last public revenue figure was $30 to $40 million. Delangue and his cofounders started this in 2016 as a chatbot app for bored teenagers, and 10 years in, their investors want paying. A rich owner might be the only thing keeping the Hub free. Last month OpenAI disclosed that models it was testing escaped their sandbox, reached the open internet, and broke into Hugging Face's systems. Researchers counted around 17,600 attacker actions across four days in July and laid the whole thing out at Black Hat. Nothing is signed. No buyer has been named. Hugging Face hasn't said a word in public. But the bank is already making calls. Every AI lab on Earth paid to keep this thing neutral. One of them is about to own it.

Ricardo

12,345 次观看 • 2 个月前

🔥NVIDIA CEO on Hugging Face Acquisition: It is “WORTH EVERY SINGLE PENNY.” $NVDA's Huang said the $12.9 billion deal was necessary to secure Hugging Face’s decade-long vision and support the rapidly growing open-source AI ecosystem.

🔥NVIDIA CEO on Hugging Face Acquisition: It is “WORTH EVERY SINGLE PENNY.” $NVDA's Huang said the $12.9 billion deal was necessary to secure Hugging Face’s decade-long vision and support the rapidly growing open-source AI ecosystem.

Coin Bureau

26,262 次观看 • 2 个月前

$NVDA understands the need to democratize AI at scale. Bringing down the cost per correct response rapidly leveraging a diverse set of open weight models. The Hugging Face deal makes enormous sense & reflects a small price for NVIDIA to pay to deepen its access to HF community.💪🏻

$NVDA understands the need to democratize AI at scale. Bringing down the cost per correct response rapidly leveraging a diverse set of open weight models. The Hugging Face deal makes enormous sense & reflects a small price for NVIDIA to pay to deepen its access to HF community.💪🏻

Daniel Newman

21,576 次观看 • 2 个月前

Meet Tencent HY-MT1.5 — our latest open-source translation models, featuring 1.8B and 7B models designed for seamless on-device and cloud deployment. 🌍 The 1.8B version is already trending at #1 on Hugging Face! 🥇It’s optimized for consumer hardware, delivering 0.18s latency (50 tokens) with just a 1GB memory footprint. Meanwhile, the 7B model achieves exceptional performance, surpassing many mid-sized models and rivaling the 90th percentile of Gemini 3.0 Pro. 🚀🧠 👉🏻 Try it now: 🔗 GitHub: 🤗 Hugging Face:

Meet Tencent HY-MT1.5 — our latest open-source translation models, featuring 1.8B and 7B models designed for seamless on-device and cloud deployment. 🌍 The 1.8B version is already trending at #1 on Hugging Face! 🥇It’s optimized for consumer hardware, delivering 0.18s latency (50 tokens) with just a 1GB memory footprint. Meanwhile, the 7B model achieves exceptional performance, surpassing many mid-sized models and rivaling the 90th percentile of Gemini 3.0 Pro. 🚀🧠 👉🏻 Try it now: 🔗 GitHub: 🤗 Hugging Face:

Tencent HY

17,105 次观看 • 10 个月前