Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Model maker John Duncan’s resin cast N-1 Starfighter maquette based on Doug Chiang’s design for Star Wars: Episode I - The Phantom Menace (1999). Chiang presented this model to George Lucas for approval (see video), while others were used as the basis of CG and physical models.

DesigningFilm's profile picture

Designing Film

6,996 subscribers

44,098 görüntüleme • 1 yıl önce •via X (Twitter)

SanatBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

9 Yorum

Designing Film profil fotoğrafı
Designing Film1 yıl önce

"George would talk about what it wanted to generally look like, what it did and how it should "feel". Then Doug and/or other artists would draw sketches to flesh out the design. Mostly the drawings were simple sketches, sometimes with close-up sketches of critical detail areas."

Designing Film profil fotoğrafı
Designing Film1 yıl önce

"Once the drawings were approved by George, they were handed off to me to do a quick, rough, small model in foam and cardboard. After the rough models were approved, I would make the finished model. But even then the design might keep evolving.”

Designing Film profil fotoğrafı
Designing Film1 yıl önce

This amazing thread on The RPF has much more detailed information and images about the models including further comments from Duncan himself:

John W.F. Hughston profil fotoğrafı
John W.F. Hughston1 yıl önce

Reminds me of the Swordfish II from Cowboy Bebop (1998)

KwakWack profil fotoğrafı
KwakWack1 yıl önce

The N-1's design itself is a masterpiece

Drex♠️ profil fotoğrafı
Drex♠️1 yıl önce

The Naboo Star-fighter is top 5 most beautiful ships in the franchise.

Ben Wiles profil fotoğrafı
Ben Wiles1 yıl önce

An iconic design. I have the lego version!

chrismitchells profil fotoğrafı
chrismitchells1 yıl önce

im a member of a facebook group that would love to see the video an photos , great photos an video

Sergorn Dragon 💉💉💉💉🌈 #NouveauFrontPopulaire profil fotoğrafı
Sergorn Dragon 💉💉💉💉🌈 #NouveauFrontPopulaire1 yıl önce

The best Star Wars ship 💓

Benzer Videolar

George Lucas personally financed the $115M budget for Star Wars: Episode I (1999). This unprecedented financial move gave him total creative control and allowed him to keep the lion's share of the film’s massive $1.027B global gross.
3:43

George Lucas personally financed the $115M budget for Star Wars: Episode I (1999). This unprecedented financial move gave him total creative control and allowed him to keep the lion's share of the film’s massive $1.027B global gross.

Best Movie Moments 🍿

587,189 görüntüleme • 5 ay önce

Star Wars Galactic Racer in VR is the best non-simulation VR driving experience ever seen in Unreal Engine; the sense of speed, the fantastic and breathtaking views of the planets in the Star Wars galaxy will bring tears of emotion to my eyes. I've been waiting for this moment since 1999 when I saw Episode I: The Phantom Menace. Thank you, praydog, for your gift! full video here:
2:04

Star Wars Galactic Racer in VR is the best non-simulation VR driving experience ever seen in Unreal Engine; the sense of speed, the fantastic and breathtaking views of the planets in the Star Wars galaxy will bring tears of emotion to my eyes. I've been waiting for this moment since 1999 when I saw Episode I: The Phantom Menace. Thank you, praydog, for your gift! full video here:

JShodanVR ➡️ SimRacingExpo 2026

451,944 görüntüleme • 7 gün önce

George Lucas foreshadowed Anakin Skywalker becoming Darth Vader in one of the smartest ways possible… In Star Wars: Episode I – The Phantom Menace, young Anakin’s shadow on the wall briefly takes the shape of Darth Vader—quietly hinting at the tragic future waiting for him long before he ever turned to the dark side.
0:22

George Lucas foreshadowed Anakin Skywalker becoming Darth Vader in one of the smartest ways possible… In Star Wars: Episode I – The Phantom Menace, young Anakin’s shadow on the wall briefly takes the shape of Darth Vader—quietly hinting at the tragic future waiting for him long before he ever turned to the dark side.

The Dark Side

958,777 görüntüleme • 5 ay önce

After the mixed critical response to Star Wars: Episode I – The Phantom Menace (1999), George Lucas was hesitant to return to the writing desk. He didn’t complete the rough draft for Attack of the Clones until March 2000, just three months before principal photography began. Lucas then continued revising the script through multiple drafts, bringing in Jonathan Hales, who had previously written episodes of The Young Indiana Jones Chronicles, to help shape the third draft into the shooting script. The final screenplay was completed just one week before filming began, making this the third Star Wars film for which Lucas shared screenplay credit, after The Empire Strikes Back and Return of the Jedi.
4:04

After the mixed critical response to Star Wars: Episode I – The Phantom Menace (1999), George Lucas was hesitant to return to the writing desk. He didn’t complete the rough draft for Attack of the Clones until March 2000, just three months before principal photography began. Lucas then continued revising the script through multiple drafts, bringing in Jonathan Hales, who had previously written episodes of The Young Indiana Jones Chronicles, to help shape the third draft into the shooting script. The final screenplay was completed just one week before filming began, making this the third Star Wars film for which Lucas shared screenplay credit, after The Empire Strikes Back and Return of the Jedi.

Best of Star Wars

28,601 görüntüleme • 3 ay önce

FLIPZONES are commonly used entry model by most SMC traders... But do you know there are types of flip zones? In this short video, i explained the 2 types of flip zones. Same flipzone entry model is what I used to catch a 1:5R on $EURUSD and $GBPUSD. Like and repost for others👍
10:02

FLIPZONES are commonly used entry model by most SMC traders... But do you know there are types of flip zones? In this short video, i explained the 2 types of flip zones. Same flipzone entry model is what I used to catch a 1:5R on $EURUSD and $GBPUSD. Like and repost for others👍

Aart 📈

69,416 görüntüleme • 2 yıl önce

Although we’ve (rightfully) become accustomed to seeing a fully CG Yoda in the Prequel Trilogy, there’s still something so special about the Yoda puppet originally used in The Phantom Menace. The technological breakthroughs made throughout this trilogy are still nothing short of incredible, but I feel like Frank Oz’s original puppeteering adds a layer of emotion and sincerity for all of Yoda’s scenes in Episode I that ultimately greatly benefit the final film. Would love to see a restoration of the theatrical version of this film from 1999 one of these days.
0:48

Although we’ve (rightfully) become accustomed to seeing a fully CG Yoda in the Prequel Trilogy, there’s still something so special about the Yoda puppet originally used in The Phantom Menace. The technological breakthroughs made throughout this trilogy are still nothing short of incredible, but I feel like Frank Oz’s original puppeteering adds a layer of emotion and sincerity for all of Yoda’s scenes in Episode I that ultimately greatly benefit the final film. Would love to see a restoration of the theatrical version of this film from 1999 one of these days.

jacob

63,494 görüntüleme • 7 ay önce

George Lucas was so impressed by Frank Oz's performance as Yoda in "Star Wars: Episode V - The Empire Strikes Back" (1980). So, he spent thousands of dollars on an advertising campaign to try and get him an Oscar nomination for "Best Actor in a Supporting Role". George Lucas's campaign ultimately failed because it was felt that a puppeteer wasn't an actor. Lucas felt this wasn't fair to Oz, who honestly didn't care. Lucas said, "After Frank Oz did "Star Wars: Episode V - The Empire Strikes Back" (1980), I tried to get him nominated for an Academy Award; but we heard back that puppetry wasn't an art. I think it is an art — and Yoda represents the highest level of that art. Acting is acting. Whether it is a human actor, a CG character, or a puppet. It's all the same. Most people think of Yoda as being real, because he is the height of puppet artistry." (Source: "Why The Oscars Refused To Nominate Star Wars' Frank Oz For His Yoda Performance", Rafael Motamayor, Slashfilm, 2023 & IMDb) P.S: On this day, 46 years ago, Irvin Kershner's "Star Wars: Episode V - The Empire Strikes Back" (1980) premiered in Washington, D.C, USA.
1:17

George Lucas was so impressed by Frank Oz's performance as Yoda in "Star Wars: Episode V - The Empire Strikes Back" (1980). So, he spent thousands of dollars on an advertising campaign to try and get him an Oscar nomination for "Best Actor in a Supporting Role". George Lucas's campaign ultimately failed because it was felt that a puppeteer wasn't an actor. Lucas felt this wasn't fair to Oz, who honestly didn't care. Lucas said, "After Frank Oz did "Star Wars: Episode V - The Empire Strikes Back" (1980), I tried to get him nominated for an Academy Award; but we heard back that puppetry wasn't an art. I think it is an art — and Yoda represents the highest level of that art. Acting is acting. Whether it is a human actor, a CG character, or a puppet. It's all the same. Most people think of Yoda as being real, because he is the height of puppet artistry." (Source: "Why The Oscars Refused To Nominate Star Wars' Frank Oz For His Yoda Performance", Rafael Motamayor, Slashfilm, 2023 & IMDb) P.S: On this day, 46 years ago, Irvin Kershner's "Star Wars: Episode V - The Empire Strikes Back" (1980) premiered in Washington, D.C, USA.

DepressedBergman

85,896 görüntüleme • 4 ay önce

🔥Really excited to see the release of PAN world model, a project I had been working over the past years. PAN is a general world model capable of simulating physical, agentic, and nested worlds, synthesizing infinite interactive experiences for training AI agents. Building on top of pretrained LLMs and video diffusion models, PAN connects language, perception, action, and latent thoughts, for long-horizon simulation and reasoning. PAN shows overwhelming performance gains over JEPA-2, Cosmos-2, and other prior models. More in the thread👇 ... 1/
0:54

🔥Really excited to see the release of PAN world model, a project I had been working over the past years. PAN is a general world model capable of simulating physical, agentic, and nested worlds, synthesizing infinite interactive experiences for training AI agents. Building on top of pretrained LLMs and video diffusion models, PAN connects language, perception, action, and latent thoughts, for long-horizon simulation and reasoning. PAN shows overwhelming performance gains over JEPA-2, Cosmos-2, and other prior models. More in the thread👇 ... 1/

Zhiting Hu

31,265 görüntüleme • 10 ay önce

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:
0:20

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 görüntüleme • 2 yıl önce

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.
0:26

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 görüntüleme • 2 ay önce

Prior to the release of "Star Wars: Episode IV - A New Hope" (1977), George Lucas screened a rough cut of the movie which was "mostly filled with stock footage of old war movies, and all kinds of stuff" to some of his friends in the film industry. Most of them thought it was a disaster. Brian De Palma was especially sarcastic in a good-natured way, teasing Lucas about the “almighty Force” and indicating that the rough cut version of the film was one of the "worst things" he had ever seen. Even Lucas thought the film was not going to be successful. He said, “I figured, well, it’s just a silly movie. It ain’t going to work.”" But Steven Spielberg was among the two who thought otherwise. He said, ‘This is going to be the biggest movie of all time’. Everybody in the room looked up at him and said, ‘Poor Steven.'” Jay cocks was the other one who praised the movie. However, Lucas' filmmaker friends helped him fix up "Star Wars" even though "they didn't have any faith in it." ("Skywalking: The Life and Films of George Lucas", Dale Pollock, 1999 & "George Lucas was convinced 'Star Wars' would flop and refused to believe it was a hit until he got a call telling him to turn on the news", kirsten Acuna, Business Insider, 2021 ) P.S: On this day, 49 years ago, "Star Wars: Episode IV - A New Hope" (1977) was released in the USA.
1:45

Prior to the release of "Star Wars: Episode IV - A New Hope" (1977), George Lucas screened a rough cut of the movie which was "mostly filled with stock footage of old war movies, and all kinds of stuff" to some of his friends in the film industry. Most of them thought it was a disaster. Brian De Palma was especially sarcastic in a good-natured way, teasing Lucas about the “almighty Force” and indicating that the rough cut version of the film was one of the "worst things" he had ever seen. Even Lucas thought the film was not going to be successful. He said, “I figured, well, it’s just a silly movie. It ain’t going to work.”" But Steven Spielberg was among the two who thought otherwise. He said, ‘This is going to be the biggest movie of all time’. Everybody in the room looked up at him and said, ‘Poor Steven.'” Jay cocks was the other one who praised the movie. However, Lucas' filmmaker friends helped him fix up "Star Wars" even though "they didn't have any faith in it." ("Skywalking: The Life and Films of George Lucas", Dale Pollock, 1999 & "George Lucas was convinced 'Star Wars' would flop and refused to believe it was a hit until he got a call telling him to turn on the news", kirsten Acuna, Business Insider, 2021 ) P.S: On this day, 49 years ago, "Star Wars: Episode IV - A New Hope" (1977) was released in the USA.

DepressedBergman

68,025 görüntüleme • 4 ay önce

Revolutionizing Move Programming with OpenLedger In this demo, we showcase how Move datasets contributed by data providers to OpenLedger’s datanets are used to fine-tune specialized models with LoRA fine-tuning. As seen in the video, we showcase an example on how builders can deploy a Move-specialized model that powers Co-pilot agents using our no-code model fine-tuning platform. This is the future of AI and Web3 innovation. Watch this space to see more specialised models and data feeds being built for next generation agents on top of OpenLedger #Move
3:58

Revolutionizing Move Programming with OpenLedger In this demo, we showcase how Move datasets contributed by data providers to OpenLedger’s datanets are used to fine-tune specialized models with LoRA fine-tuning. As seen in the video, we showcase an example on how builders can deploy a Move-specialized model that powers Co-pilot agents using our no-code model fine-tuning platform. This is the future of AI and Web3 innovation. Watch this space to see more specialised models and data feeds being built for next generation agents on top of OpenLedger #Move

OpenLedger

61,697 görüntüleme • 1 yıl önce

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.
4:10

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.

AK

290,517 görüntüleme • 3 yıl önce

I wasn’t planning to go over budget, then I realized this estate belonged to a Thunderbirds pilot… Last night, I was scrolling estate sales to make sure I hadn’t missed anything for the weekend. One listing had just added photos, and it was well worth the wait. The estate belonged to Robert Jackson, a former USAF Thunderbirds pilot who flew left wing from 1970–71 during the F-4 Phantom era. The first thing that stopped me mid-scroll was the Toppings models. I went a bit over budget, but walked out with some incredible finds: 1.Toppings YF-16 model AF 01-568 in two-tone Blue (damaged but still beautiful) 2.Toppings Blue Angels F-4 Phantom model 3.A-7D Corsair model 4.Toppings T-38 model 91605 5.Toppings Lockheed F-104 Starfighter model 6.Thunderbirds Wives Cookbook, a surprisingly special piece featuring recipes from the wives of the team at the time There was one more item I picked up, but I’ll have to wait until tomorrow to share it. It’s a gift for my husband.
0:34

I wasn’t planning to go over budget, then I realized this estate belonged to a Thunderbirds pilot… Last night, I was scrolling estate sales to make sure I hadn’t missed anything for the weekend. One listing had just added photos, and it was well worth the wait. The estate belonged to Robert Jackson, a former USAF Thunderbirds pilot who flew left wing from 1970–71 during the F-4 Phantom era. The first thing that stopped me mid-scroll was the Toppings models. I went a bit over budget, but walked out with some incredible finds: 1.Toppings YF-16 model AF 01-568 in two-tone Blue (damaged but still beautiful) 2.Toppings Blue Angels F-4 Phantom model 3.A-7D Corsair model 4.Toppings T-38 model 91605 5.Toppings Lockheed F-104 Starfighter model 6.Thunderbirds Wives Cookbook, a surprisingly special piece featuring recipes from the wives of the team at the time There was one more item I picked up, but I’ll have to wait until tomorrow to share it. It’s a gift for my husband.

Tori Mae Fontana

16,306 görüntüleme • 7 ay önce

Had a chance this morning to drive the NEW Model Y AND do a short FSD v13 video with the car. The car drives incredible and FSD works as good as ever! To my knowledge, this is the first known video of FSD being used on the New Model Y! I also did some manual driving & tested out the reduced regen mode for the breaking which will be great for those Uber drivers who can't get one pedal driving down. It worked fantastic, and the car drove great coming from the Model X! Sorry for the shaky cam in-advance. Couldn't use the Go Pro.
8:46

Had a chance this morning to drive the NEW Model Y AND do a short FSD v13 video with the car. The car drives incredible and FSD works as good as ever! To my knowledge, this is the first known video of FSD being used on the New Model Y! I also did some manual driving & tested out the reduced regen mode for the breaking which will be great for those Uber drivers who can't get one pedal driving down. It worked fantastic, and the car drove great coming from the Model X! Sorry for the shaky cam in-advance. Couldn't use the Go Pro.

Jordan Camina

298,224 görüntüleme • 1 yıl önce

A breakthrough in real-time video generation. As a research preview developed with NVIDIA and shared at NVIDIAGTC this week, we trained a new real-time video model running on Vera Rubin. HD videos generate instantly, with time-to-first-frame under 100ms. Unlocking an entirely new creative paradigm and bolstering the foundations of our General World Model, GWM-1. Real-time generation opens a fundamentally different design space for video models and world simulation. We're investing in co-designing our models alongside advances in hardware to keep pushing this frontier.
0:54

A breakthrough in real-time video generation. As a research preview developed with NVIDIA and shared at NVIDIAGTC this week, we trained a new real-time video model running on Vera Rubin. HD videos generate instantly, with time-to-first-frame under 100ms. Unlocking an entirely new creative paradigm and bolstering the foundations of our General World Model, GWM-1. Real-time generation opens a fundamentally different design space for video models and world simulation. We're investing in co-designing our models alongside advances in hardware to keep pushing this frontier.

Runway

1,165,999 görüntüleme • 6 ay önce

🌎🇷🇺"Global development model was based around a restricted number of financial centers, technological solution, insurance & logistics hubs, rating agencies, as well as reserve currencies. This architecture was portrayed as "universal". It was increasingly often used as an instrument for political pressure and unfair competition to punish those who choose to act in their own national interest." — Putin on the old development model of globalists
1:03

🌎🇷🇺"Global development model was based around a restricted number of financial centers, technological solution, insurance & logistics hubs, rating agencies, as well as reserve currencies. This architecture was portrayed as "universal". It was increasingly often used as an instrument for political pressure and unfair competition to punish those who choose to act in their own national interest." — Putin on the old development model of globalists

Lord Bebo

15,839 görüntüleme • 4 ay önce

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.
1:43

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

AK

40,492 görüntüleme • 2 yıl önce

Nvidia has just announced Alpamayo 2 Super, an open 34 billion parameter reasoning vision-language-action model designed to accelerate the development of autonomous vehicles. This new model combines the NVIDIA Cosmos 3 Super reasoning model with a 2 billion parameter diffusion-based action expert model, and is post trained with reinforcement learning. The model can return multiple outputs: future trajectory plans, reasoning traces, grounded answers to questions about the scenes, and auto label generation. The model weights are now available for anyone to download on Hugging Face, and the inference code has been posted to GitHub. Distilled models can be deployed commercially without any further permission from Nvidia, and model outputs carry no license conditions. Automakers can distill down a compact version of this model that can run on the Nvidia computer in the car. Major kudos to Nvidia and Jensen Huang for advancing the state of the industry by releasing this as an open model with permissive licensing. Jensen isn't just paying lip service to the idea of open models, Nvidia is actually contributing to the ecosystem — and it's great for their business, because it helps sell more Thor computers that go in the car. Anyone can go download the model and play with it. If you do, let me know what you think. Personally I think it's so cool that we have open weights models that are this advanced, for anyone to download.
2:30

Nvidia has just announced Alpamayo 2 Super, an open 34 billion parameter reasoning vision-language-action model designed to accelerate the development of autonomous vehicles. This new model combines the NVIDIA Cosmos 3 Super reasoning model with a 2 billion parameter diffusion-based action expert model, and is post trained with reinforcement learning. The model can return multiple outputs: future trajectory plans, reasoning traces, grounded answers to questions about the scenes, and auto label generation. The model weights are now available for anyone to download on Hugging Face, and the inference code has been posted to GitHub. Distilled models can be deployed commercially without any further permission from Nvidia, and model outputs carry no license conditions. Automakers can distill down a compact version of this model that can run on the Nvidia computer in the car. Major kudos to Nvidia and Jensen Huang for advancing the state of the industry by releasing this as an open model with permissive licensing. Jensen isn't just paying lip service to the idea of open models, Nvidia is actually contributing to the ecosystem — and it's great for their business, because it helps sell more Thor computers that go in the car. Anyone can go download the model and play with it. If you do, let me know what you think. Personally I think it's so cool that we have open weights models that are this advanced, for anyone to download.

Whole Mars Catalog

45,595 görüntüleme • 2 ay önce