SANTA CLARA — At the GTC conference, NVIDIA unveiled a comprehensive catalog of enterprise-grade generative AI microservices designed to empower businesses to develop and deploy custom applications while retaining control over their intellectual property. Leveraging the NVIDIA CUDA® platform, these cloud-native microservices offer optimized inference capabilities across various domains, including language processing, drug discovery, and high-performance computing. Notable among these microservices are the NVIDIA NIM microservices, which enable developers to reduce deployment times from weeks to minutes, and the CUDA-X™ microservices, which provide end-to-end building blocks for data processing, customization, and training. These microservices, adopted by leading application providers such as Adobe, SAP, and ServiceNow, mark a significant advancement in enterprise AI, offering standardized solutions for leveraging NVIDIA’s vast installed base of GPUs across clouds, data centers, workstations, and PCs.
NVIDIA’s NIM microservices are poised to accelerate the deployment of AI models in production environments, providing pre-built containers powered by NVIDIA inference software. These microservices offer industry-standard APIs for domains such as language, speech, and drug discovery, enabling developers to build AI applications quickly and securely. Moreover, enterprises can access these microservices from popular cloud platforms and integrate them with AI frameworks like Deepset and LlamaIndex, facilitating seamless integration into existing workflows.
Meanwhile, NVIDIA’s CUDA-X microservices offer a range of capabilities, from data preparation and customization to routing optimization and high-resolution climate simulations. Notably, the NeMo Retriever™ microservices empower developers to link AI applications with business data, enhancing the accuracy and relevance of responses. Additionally, NVIDIA’s ecosystem of partners, including leading data platform providers like Cloudera and infrastructure providers like Dell Technologies, is collaborating to optimize RAG pipelines and integrate proprietary data into generative AI applications.
Developers can experiment with NVIDIA microservices at ai.nvidia.com, while enterprises can deploy production-grade NIM microservices with NVIDIA AI Enterprise 5.0 on certified systems and leading cloud platforms. NVIDIA’s commitment to democratizing AI through accessible and scalable solutions promises to accelerate innovation across industries, from healthcare to cybersecurity and beyond. For more information, attendees can visit NVIDIA’s booth at GTC or watch the replay of Jensen Huang’s keynote address.
