NVIDIA Video Super Resolution gRPC NIM is the API-driven flavor for GPU-accelerated video upscaling over gRPC and streaming-protocol workflows.


NVIDIA Video Super Resolution gRPC NIM
What Is NVIDIA NIM?
NVIDIA NIM packages AI inference services as containerized, GPU-accelerated microservices. NIM containers are designed to help teams deploy AI models and media processing services with documented APIs, release artifacts, and production-oriented deployment guidance.
The NVIDIA Video Super Resolution (VSR) gRPC NIM is the gRPC flavor of the Video Super Resolution NIM release. It provides AI-powered video upscaling for GPU-accelerated video processing workflows. It enhances lower-resolution video to higher resolutions while preserving visual quality, enabling integration into video processing and media enhancement pipelines.
The container includes the NVIDIA Video Super Resolution model and runtime components needed to run Video Super Resolution inference on supported NVIDIA GPU systems through gRPC and related streaming-protocol workflows.
Use this container for the gRPC flavor of Video Super Resolution NIM. For the
SMPTE ST 2110 flavor on NVIDIA Holoscan for Media, use the vsr-h4m-nim
container.
System Requirements
For compressed-video gRPC workflows, the following are required:
- An Ada-generation or later NVIDIA GPU. One GPU is required per VSR NIM container.
- GPU NVENC support for the selected output codec and NVDEC support for the selected input codec. Refer to the NVIDIA Video Encode and Decode GPU Support Matrix to verify codec support.
- NVIDIA Driver 590.33 or later.
- CUDA 13.1 or later.
- Docker and NVIDIA Container Toolkit.
Getting Started with NVIDIA NIM
For release documentation, deployment guidance, supported workflows, and configuration details, see the Video Super Resolution gRPC NIM Documentation.
Get Help
Enterprise Support
Get access to knowledge base articles and support cases or submit a ticket.
Governing Terms
The NVIDIA Video Super Resolution NIM container is governed by the NVIDIA Software License Agreement and the Product Specific Terms for NVIDIA AI Products.
Use of the DLPP 2.1 model is governed by the NVIDIA Open Model License.