img.aligncenter {
aspect-ratio: 1.7777777777778;
margin: 0 auto;
display: block;
}

strong {
font-weight: bold;
}

Nvidia

Nvidia has just unveiled its brand-new generative AI model named Latte3D during GTC 2024. Latte3D seems to be like ChatGPT on steroids. It is a text-to-3D model that can accept simple and short text prompts and turn them into 3D objects and animals in an instant. And it’s much quicker than its predecessors. Latte3D functions like a virtual 3D printer that could be extremely useful for creators in various industries.
Latte3D is designed to simplify the creation of 3D models for different types of creators, such as those working on video games, design projects, marketing, or even for robotics-related machine learning and training. In Nvidia’s demonstration of the model, it appears incredibly easy to use. After a quick text prompt, the AI generates a 3D model and then quickly enhances it with more details. Although the final result isn’t as lifelike as OpenAI’s Sora, that’s not its purpose. This is a way to speed up the creation of assets instead of having to build them from scratch.
The model offers several options for the user to choose from, and Nvidia claims that these shapes can be “optimized for higher quality within a few minutes.” The designed models can then be exported to different platforms, like Nvidia’s Omniverse, and can be adjusted to match the desired outcome. Nvidia trained Latte3D using its Ada A100 Tensor Core GPUs and supported the training with ChatGPT prompts to prepare it for interacting with real users.
Currently, Latte3D can only generate objects and animals. In this regard, it seems to perform a solid job in distinguishing different animals, textures, and object types. Nvidia showcased these capabilities by presenting objects like an amigurumi (crochet) common crane or an origami sphynx cat. The model has been taught to recognize various species, so it can tell the difference between an Italian greyhound and a Shiba Inu.
Creators who want to utilize Latte3D for more advanced tasks can train it on a different dataset, whether it’s plants or household objects, and then use it for their own purposes. Nvidia brings up some interesting use cases here, such as training personal assistant robots before deploying them. It’s easy to envision that Latte3D will be extremely handy for game developers, but its potential goes far beyond just gaming scenarios.
Sanja Fidler, the vice president of AI research at Nvidia, remarked on how much faster Latte3D is compared to its predecessors. “A year ago, it took an hour for AI models to generate 3D visuals of this quality – and the current state of the art is now around 10 to 12 seconds. We can now produce results an order of magnitude faster,” said Fidler.
The recent announcements related to the use of AI in game development are all quite groundbreaking, and Nvidia’s Latte3D joins a growing list of tools that may one day completely transform the process of creating a game. For instance, Nvidia just recently unveiled non-player characters (NPCs) with dialogue entirely generated by AI. Meanwhile, Unreal Engine’s latest update can generate film-quality visuals in games in real time, all with the help of machine learning.

Get your weekly teardown of the tech behind PC gaming
ReSpec
Subscribe
Check your inbox!
Privacy Policy

As of now, Latte3D can only generate objects and animals. In this regard, it seems to do a commendable job in discerning different animals, textures, and object types. Nvidia demonstrated these capabilities by presenting objects such as an amigurumi (crochet) common crane or an origami sphynx cat. The model has been trained to recognize various species, thus enabling it to distinguish between an Italian greyhound and a Shiba Inu.
Creators who wish to utilize Latte3D for more extensive applications can train it on a different dataset, whether it’s related to plants or household objects, and then utilize it for their own purposes. Nvidia presents some interesting use cases here, such as training personal assistant robots before deploying them. It’s easy to imagine that Latte3D will be highly beneficial for game developers, but its potential extends far beyond just gaming scenarios.
Sanja Fidler, the vice president of AI research at Nvidia, commented on how much faster Latte3D is compared to its predecessors. “A year ago, it took an hour for AI models to generate 3D visuals of this quality – and the current state of the art is now around 10 to 12 seconds. We can now achieve results at a speed that is an order of magnitude faster,” said Fidler.
The recent announcements regarding the use of AI in game development are all extremely innovative, and Nvidia’s Latte3D joins a growing list of tools that could potentially revolutionize the process of creating a game. For instance, Nvidia just recently unveiled non-player characters (NPCs) with dialogue entirely generated by AI. Meanwhile, Unreal Engine’s latest update can generate visuals in games with film-quality in real time, all with the assistance of machine learning.

LATTE3D Text to 3D Generative AI Model from NVIDIA Research

Creators who want to use Latte3D to do more can train it on a different dataset, be it plants or household objects, and later use it for their own purposes. Nvidia brings up some interesting use cases here, such as training personal assistant robots before deploying them. It’s easy to imagine that Latte3D will come in handy for game devs, but the potential goes far beyond just gaming scenarios.
Sanja Fidler, vice president of AI research at Nvidia, remarked on how much faster Latte3D is compared to its predecessors: “A year ago, it took an hour for AI models to generate 3D visuals of this quality — and the current state of the art is now around 10 to 12 seconds. We can now produce results an order of magnitude faster,” said Fidler.
The recent announcements related to using AI in game development are all pretty groundbreaking, and Nvidia’s Latte3D joins a growing list of tools that may one day completely change the process of creating a game. For instance, Nvidia just recently unveiled non-player characters (NPCs) with dialogue entirely generated by AI. Meanwhile, Unreal Engine’s latest update can generate film-quality visuals in games in real time, all with the help of machine learning.