Support Matrix#
Models#
Model Name |
Model ID |
Publisher |
|---|---|---|
Relighting |
ai4m-relighting |
NVIDIA |
Optimized Configurations#
Server GPU |
Precision |
|---|---|
T4 |
FP16 |
A10, A40, A100 |
FP16 |
L4, L40, L40S |
FP16 |
H100 |
FP16 |
B40, BA100, BA200 |
FP16 |
Other Architectures (Consumer and Desktop GPUs)#
Consumer GPU |
Precision |
|---|---|
RTX 4090 |
FP16 |
RTX 5080, RTX 5090 |
FP16 |
The Relighting NIM is compatible with professional, datacenter, and consumer GPUs that have Tensor cores and are based on the following NVIDIA architectures: Blackwell, Ada, Ampere, Hopper, and Turing.
The NIM requires NVENC/NVDEC hardware for video encoding/decoding. For details about supported GPUs and H264 YUV formats, refer to the Video Encode and Decode Support Matrix.
Software#
NVIDIA Driver and Prerequisites#
NVIDIA driver requirements and other prerequisites for Relighting NIM:
Prerequisite |
Version |
Download and install steps |
|---|---|---|
NVIDIA Graphic Drivers for Linux |
571.21+ |
|
Docker |
latest |
Ubuntu, CentOS, and Debian: https://docs.docker.com/engine/install/; Rocky Linux: https://docs.rockylinux.org/gemstones/containers/docker/ |
NVIDIA Container Toolkit |
latest |
Installation and configuration instructions: https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html |
Relighting NIM uses the following NVIDIA software platforms:
Components |
Version |
|---|---|
CUDA |
12.8.1 |
cuDNN |
9.7.1.26 |
TRT |
10.9.0.34 |
Triton Inference Server |
v2.50.0 |
DeepStream |
8.0 |