NVIDIA ships BioNeMo Inference Runtime v0.2 with RFD4-Proteina support
AllThingsApx · x · 2026-10-24
NVIDIA Health released BioNeMo Inference Runtime (BioIR) v0.2, adding support for the newly open-sourced RFD4-Proteina and delivering broad performance gains. Installable via pip (bionemo-ir 0.2.0).
Key points:
- A five-stage GPU pipeline turns AlphaFold-lineage and all-atom models from FASTA/MSA input into PDB/mmCIF structures with confidence scores
- Models stay ordinary nn.Modules — no TensorRT engine build required
- Beta caveats: needs Linux x8664/aarch64, NVIDIA GPU, driver 580+, Python 3.12, glibc 2.34+; running via the forward-compatibility shim on older drivers can hang or crash, and results are discarded
More from Infra
- What $70-100K of hardware buys for local GLM5.3 and DeepSeek with 10-20 concurrent users — whoami-233 · 2026-10-24
- Musk says no H100s to rent; experts urge banning Chinese firms from renting US AI chips — chrisrohlf · 2026-10-24
- Amazon scraps Northern Virginia data center plans after local opposition, donates land — Polymarket · 2026-10-24
- Stripe reportedly buying OpenRouter for ~$7.5B as AI usage billing becomes the prize — FinanceYF5 · 2026-10-24
- Redditor pushes back on local AI FUD: real inference speeds are 5x what the video shows — wojtek15 · 2026-10-24
- ISTA-DASLab ships GSQ-RCO quantized DeepSeek-V4.1-Flash GGUFs, llama.cpp support still pending — Gold_Coconut9777 · 2026-10-24