NVIDIA Launches AI Foundations – Services For Creating Unique Chat Bots, Picture Generators And Other AI Tools
At a presentation at the GTC conference, Nvidia announced the AI Foundations cloud service. It is aimed at companies that are interested in working with artificial intelligence, but do not have sufficient resources.
With the help of AI Foundations, companies will be able to create their own models that run on the Nvidia infrastructure and use the company’s technologies. For example, AI Foundations users will be able to train their version of an Nvidia chatbot by providing training data.
Also, users of the service will have access to the NeMo neural network, which generates an image according to the description, and Picasso technology, which can create not only images and videos, but also three-dimensional applications.
Adobe, Getty Images, Morningstar, Quantiphi and Shutterstock are among the companies using NVIDIA cloud services.
Each cloud service includes six elements: pre-trained models, data processing platforms, vector databases and personalization, optimized inference engines, APIs, and support from NVIDIA experts in model tuning.
The NeMo cloud service allows developers to tailor large language models (LLMs) to their needs by prioritizing, adding domain knowledge, and learning the necessary functional skills. Models ranging in size from 8 to 530 billion parameters are available in the service with regular updates, which gives enterprises ample opportunity to create applications that meet business requirements for speed, accuracy and complexity of tasks.
Using the search capabilities included with the NeMo service, customers will be able to supplement LLM with their own real-time data. This allows businesses to customize models for market intelligence, enterprise search, chatbots, and customer service.
Generative AI cloud service NeMo is still in early access, while Picasso is in closed preview. Developers can apply for access to each service through the appropriate links on the NVIDIA website.
