NIM Microservices

NVIDIA NIM™ provides prebuilt, optimized inference microservices that let you deploy the latest AI foundation models with security and stability on any NVIDIA-accelerated infrastructure. They are provided as downloadable containers which are built following all the procedures outlined earlier in this document. NIM microservices expose industry-standard APIs for simple integration into AI applications, development frameworks, and workflows. Access to download all NIM containers is included with the NVIDIA AI Enterprise license.

NVIDIA performs additional security measures for NIM microservices to provide customers with a way to run any supported model in their own infrastructure.

Licensing & Compliance

For every model delivered as an NVIDIA Community Model, NVIDIA evaluates the source code, including downstream dependencies for licenses and distribution rights, identical to any other open-source software. In addition, NVIDIA evaluates the license of the model weights as well.

NVIDIA participates in and contributes to various open-source communities. When shipping a new NIM, the team conducts a thorough license check for code and dependencies.

Code Audit / Review

Once software has been published, it must be maintained, and this includes addressing any vulnerabilities that may be discovered after publication. In this section, we will go over how NVIDIA helps customers maintain the security of our software after deployment.