NVIDIA Launches AI Microservices to Help Developers Build Custom AI Apps

NVIDIA launched AI microservices for developers to create custom generative AI applications on their platforms while retaining IP ownership.

The microservices are optimized to run on NVIDIA's CUDA platform across clouds, data centers, workstations and PCs.

They include NIM for optimized inference and CUDA-X for capabilities like retrieval-augmented generation, guardrails, and data processing.

Leading platforms like Adobe, Cadence, CrowdStrike, SAP, and ServiceNow are adopting the new microservices.

The microservices connect AI model developers, platform providers, and enterprises to NVIDIA's CUDA ecosystem of hundreds of millions of GPUs.