Supported Hardware | NVIDIA BioNeMo NIM Documentation

The OpenFold3 NIM is configured to run on a single GPU, and supports different sequence length ranges depending on the inference backend:

PyTorch backend: Supports sequences from 2 to 4096 amino acids
TensorRT-BioNeMo backend: Supports sequences from 4 to 2048 amino acids (optimized for faster inference)

GPU

The OpenFold3 NIM is supported to run on the GPU configurations below.

GPU	GPU Memory	# of GPUs
NVIDIA H100 80GB HBM3	80GB	[1]
NVIDIA H100 PCIe	80GB	[1]
NVIDIA A100-SXM4-80GB	80GB	[1]
NVIDIA A100 80GB PCIe	80GB	[1]
NVIDIA H200	141GB	[1]
NVIDIA B200	180GB	[1]
NVIDIA GB200	186GB	[1]
NVIDIA L40S	48GB	[1]
NVIDIA GB10 (DGX Spark)	128GB	[1]

The GB10 features 128GB of LPDDR5x unified memory shared between CPU and GPU, providing seamless memory access across the Grace Blackwell Superchip architecture.

Storage

NVIDIA recommends at least 80GB of disk space. The container is roughly 61GB, which includes the OpenFold3 model parameters (~6GB) and TensorRT optimized inference engines.

CPU and System Memory

At least 64 GB of RAM
A CPU with at least 8 available cores