NVIDIA has launched NeMo Retriever, an AI microservice, and is collaborating with Cadence, Dropbox, SAP, and ServiceNow to implement it. This service enhances natural language search capabilities and enables more accurate AI inference applications, such as chatbots, Copilot assistance features, and content summarization tools.
NeMo Retriever is a new addition to the NVIDIA NeMo family of frameworks and tools for building, customizing, and deploying generative AI models. It helps organizations improve the effectiveness of automated generative AI applications through enterprise-grade Retrieval Augmented Generation (RAG).
Through NVIDIA algorithm optimization, NeMo Retriever will enable automated AI applications to provide more accurate responses. Unlike open-source search enhancement toolkits, NeMo Retriever leverages commercially available application models, API resources, security patches, and enterprise resources to build automated AI applications. For example, it allows users to interact with data and obtain accurate and timely answers through simple conversations.
Enterprises can deploy NeMo Retriever-powered applications in any data center or on NVIDIA accelerated computing devices in the cloud for inference. Developers are now allowed to register to experience the NeMo Retriever application features first.
