Skip to navigation

Support Matrix

View as Markdown

GenMol NIM v2.0.0 uses a single generic FP32 profile and runs on one GPU per container instance.

Model Profile

PrecisionDisk SpaceProfile ID
FP32340 MB9a60e7446868a0e75a97f5aa7f9b5312bfc8f221d6855b16a54dd0064ffd77b3

GPU

GenMol NIM v2.0.0 is validated on the following GPU configurations.

GPUGPU Memory# of GPUs
NVIDIA RTX 6000 Ada Generation48 GB1
NVIDIA L40S48 GB1
NVIDIA A10G24 GB1
NVIDIA A10040 GB or 80 GB1
NVIDIA H10080 GB1
NVIDIA H200141 GB1
NVIDIA B200180 GB1
NVIDIA GB200186 GB1
NVIDIA B300288 GB1
NVIDIA GB300288 GB1
NVIDIA GB10 (DGX Spark)128 GB1
NVIDIA GH200 144GB144 GB1
NVIDIA RTX PRO 6000 Blackwell Workstation Edition96 GB1

The GB10 entry corresponds to the DGX Spark platform and uses 128 GB of unified memory shared across CPU and GPU.

Storage

Use at least 20 GB of local disk space for the container image, model assets, and cache directory.

CUDA Compute Capability

This NIM requires CUDA Compute Capability 7.0 or higher.

Supported Software