NVIDIA has introduced the Jetson Orin Nano Super Developer Kit along with NeMo Retriever suite of microservices – tools designed to enhance generative AI (GenAI) capabilities.
Nvidia, leading multinational technology company, on Dec. 17 presented two of GenAI-related innovations. These are Jetson Orin Nano Super Developer Kit, a compact generative AI supercomputer priced at $249, and NeMo Retriever models able to enhance multilingual information retrieval for generative AI applications.
The supercomputer device has advanced tech specs. It can deliver up to 67 INT8 TOPS of AI performance, a 1.7x increase over its predecessor, and features an enhanced memory bandwidth of 102GB/s. Jetson Orin Nano Super Developer Kit consists of a Jetson Orin Nano 8GB system-on-module (SoM) and a reference carrier board, making it a perfect platform for prototyping edge AI applications. It supports popular generative AI models, including large language models and vision transformers, making it accessible to developers, students, and hobbyists.
Meanwhile, NeMo Retriever services enable enterprises to connect AI models to diverse data sources across various languages, improving the accuracy and relevance of AI outputs. It allows businesses to extract insights from vast datasets, get more accurate responses, and connect generative AI to enterprise data in multiple global languages to reach broader audiences. Besides, businesses using the new tool can scale actionable intelligence with 35x greater data storage efficiency using advanced techniques like long context support and dynamic embedding sizing.
Notably, the firm revealed an outstanding use case for its new technology. In collaboration with DataStax, Wikimedia has utilised NeMo Retriever to vectorise over 10 million Wikidata entries in under three days — a process that previously took 30 days — thereby significantly enhancing data processing efficiency.
Nvidia’s innovations are often used to explore the practical application and potential of the most advanced technologies. We have previously reported that Google Quantum AI was using the Nvidia CUDA-Q platform and the Nvidia Eos supercomputer to develop its next-generation quantum computing devices.
Together, the Jetson Orin Nano Super Developer Kit and NeMo Retriever microservices provide powerful tools for developers and organisations to advance generative AI capabilities, offering scalable solutions for complex AI tasks across multiple languages and data formats.
