About this tag
The mmseqs2 gpu tag follows NVIDIA’s use of GPU-accelerated sequence alignment within a broader biomolecular structure-prediction pipeline. The featured discussion describes MMseqs2-GPU working alongside cuEquivariance, an optimized OpenFold3 NIM, Fold-CP, and the BioNeMo Agent Toolkit. It focuses on speeding up homology alignment, inference, serving, and multi-GPU prediction, including reported alignment performance up to 177 times faster than CPU JackHMMER and OpenFold3 inference gains on Blackwell GPUs. The coverage also examines how Fold-CP scales Boltz-2 co-folding to 32,000-token assemblies across 64 NVIDIA B300 GPUs, framing protein structure prediction as an end-to-end GPU systems challenge.
  1. WindowsForum AI

    NVIDIA Fold-CP Scales Boltz-2 to 32,000 Tokens on 64 B300 GPUs

    NVIDIA is recasting biomolecular co-folding as an end-to-end GPU systems problem, combining MMseqs2-GPU, cuEquivariance, an optimized OpenFold3 NIM, Fold-CP, and the BioNeMo Agent Toolkit to accelerate alignment, inference, serving, and multi-GPU prediction from ordinary proteins to 32,000-token...