NVIDIA's BioNeMo Runtime Boosts Protein Structure Prediction Throughput 2.9x
AllThingsApx · x · 2026-09-24
NVIDIA's BioNeMo Inference Runtime (BioIR) accelerates biomolecular structure-prediction models while keeping the PyTorch workflow intact.
- Benchmark: On 8xH100s over 1,000 human dimer targets, BioIR-accelerated Boltz-2 folded 58.5K residues per GPU-hour vs 20.2K for a torch-compiled open-source baseline — a 2.90x residue-normalized throughput gain.
- Three-layer optimization: kernel selection, module optimization via CUDA Graph capture, and pipeline scaling with Ray replicas (one full model replica per GPU, overlapping CPU stages with GPU folding).
- Energy: folding 1M targets is estimated at 11 MWh with BioIR vs 35 MWh for the public implementation.
- The end-to-end pipeline spans parsing, tokenization, feature generation, GPU inference, and PDB/mmCIF output, with direct PyTorch module integration. Code is open.
More from Infra
- UK Sovereign AI invests in Fractile to build next-gen frontier AI chips — charlieharris01 · 2026-11-11
- What happened when I moved vector retrieval into Snowflake SQL instead of orchestrating RAG outside — dima806_dima · 2026-11-11
- AI chip startup Fractile raises $720M Series C for DRAM-on-processor design — Tom_Westgarth15 · 2026-11-11
- Mobile AI Factory: agentic pipeline builds on-device SLMs that beat 5-shot frontier models — kalyan_kpl · 2026-11-11
- Vivo's Starlink direct-to-cell service hits 10M users in two weeks across Brazil — XFreeze · 2026-11-11
- Nan Ya Plastics to spend NT$11.53B on fifth glass fabric plant on AI server demand — zephyr_z9 · 2026-11-11