Supported Hardware#
The OpenFold2 NIM is configured to run on a single GPU, and supports input sequences with length less than (or-equal-to) 1000 AAs.
GPU#
The OpenFold2 NIM is supported to run on a single NVIDIA GPU, with at least 80GB of video RAM. GPU with 80GB video RAM are in the Ampere, Hopper, and Blackwell architectures.
Examples:
GPU |
GPU Memory |
# of GPUs |
---|---|---|
NVIDIA H100 80GB HBM3 |
80GB |
[1] |
NVIDIA A100 80GB PCIe |
80GB |
[1] |
Storage#
NVIDIA recommends at least 350GB of disk space. The container is roughly 30GB. Additionally, at startup, the NIM downloads needed data items, if they are not present on the local filesystem. These data items are model parameters (roughly 10GB), and a dataset of protein structures (roughly 300GB).
CPU and System Memory#
At least 64 GB of RAM
A CPU with at least 8 available cores