Snowflake and NVIDIA are integrating AI-powered applications, models, and hardware to help customers easily, efficiently, and securely deploy enterprise AI across their businesses.

Snowflake, a global AI data cloud company, announced at Snowflake Summit 2024, held in San Francisco, USA from June 3rd to 6th, that it is strengthening its collaboration with NVIDIA to enable companies to build customized AI data applications based on NVIDIA AI on its platform.

Through this collaboration, Snowflake has integrated the NeMo Retriever microservice into Snowflake Cortex AI, a fully managed LLM (Large-Scale Language Model) and vector search service, using NVIDIA AI Enterprise software.

This allows customers to seamlessly connect customized models to diverse business data and provide accurate responses.

Snowflake Arctic, an enterprise-grade LLM with industry-leading openness, supports NVIDIA TensorRT-LLM (TRT-LLM) software to deliver highly optimized performance.

Also NVIDIA NIM (NVIDIA Inference Microservice) is now available as an inference microservice, allowing more developers to gain insights from Arctic.

“Integrating NVIDIA’s full-stack accelerated computing and software with Snowflake’s advanced AI capabilities into Cortex AI is a game-changer,” said Sridhar Ramaswamy, CEO of Snowflake. “Together, we are working to usher in a new era of AI, where customers across all industries and skill levels can easily, efficiently, and securely build customized AI applications on their enterprise data.”

“Data is the essential commodity for the AI industrial revolution,” said Jensen Huang, founder and CEO of NVIDIA. “Our collaboration will help customers refine their enterprise-specific data and transform it into valuable, generative AI.”