Open-weight & owned stack · Quantization efficiency

NVIDIA releases BioNeMo Inference Runtime for protein folding workloads

Sep 10, 2026, first seen via MarkTechPost

NVIDIA releases BioIR inference runtime for protein folding

Read the original at marktechpost.comOpens the publisher's site in a new tab

Also covering this

NVIDIA Details BioNeMo Inference Runtime (BioIR): 2.90x Higher Boltz-2 Folding Throughput and 58.5K Residues per GPU-Hour on 8xH100 - MarkTechPostmarktechpost.com, Sep 10

More in Open-weight & owned stack

Z.ai publishes GLM-5.3 open weightsSep 8 MiniMax H3 highlighted as open-weight multimodal modelSep 7 InternLM releases 397B multimodal Intern-S2 modelSep 13 DeepSeek Model Nearly Matches GPT-6 Astra DesignSep 10 DeepSeek presents V4 FlashSep 8