Nvidia has made a long list of announcements at the GTC conference covering the full range of AI operations in numerous industries.
One of the major announcements Nvidia made was about a number of new Large Language Models (LLM) capabilities, including a pair of cloud services that are meant to make it easier for more businesses and people to develop, train, and use LLMs. The Nvidia NeMo LLM Service and the Nvidia BioNeMo LLM Service are two of the new cloud services available.
According to the company, the large AI models that the service will support include ones with 5 billion, 20 billion, and 530 billion parameters based on the Megatron LLM. Nvidia is supporting prompt learning-based tuning as part of the service to quickly enable use cases that are specific to a given context.