NVIDIA Shares Breakthroughs in Accelerating Protein Design
AllThingsApx · x · 2026-07-16
NVIDIA's Digital Biology team has shared several new acceleration milestones in protein structure prediction and design:
- GPU MSA search speed has increased by 177x.
- OpenFold3 inference is now 4x faster on NVIDIA Blackwell GPUs.
- Fold-CP can scale to complexes with 32,000 tokens on 64 NVIDIA B300 GPUs.
- These capabilities are orchestrated into agentic workflows within the BioNeMo Agent Toolkit.
The original post also mentions collaborations with multiple partners, aiming to provide a faster end-to-end biomolecular design pipeline on the open BioNeMo platform.
More from Infra
- NVIDIA publishes Vera CPU architecture details before AMD’s AI event — ryanshrout · 2026-07-22
- oMLX 0.5.2 adds Mac menu-bar stats, low-bit decode kernels, and faster downloads — awnihannun · 2026-07-22
- Strangeworks launches Aura to turn enterprise ops into production optimization systems — whurley · 2026-07-22
- Graph workload 854.graph500 enters SPEC CPU 2026 as a new CPU benchmark — Prof_DavidBader · 2026-07-22
- HilbertRaum open-sources a fully local AI chat and document analysis app for private use — Vladowski · 2026-07-22
- Hybrid and local inference are emerging as a response to AI energy and token costs — dmitry140 · 2026-07-22