AWS adds OpenFold3 to SageMaker JumpStart for private biomolecular structure prediction
AWS ML Blog · rss · 2026-11-14
AWS ML Blog announces that OpenFold3-preview is now deployable from Amazon SageMaker JumpStart. Developed by the OpenFold Consortium and Columbia's AlQuraishi Lab under Apache 2.0, it predicts all-atom 3D structures of proteins, DNA, RNA, small-molecule ligands, and their complexes from sequences (ligands as SMILES), with pLDDT/PAE/PDE confidence metrics.
Highlights
- Supports multi-chain complexes, protein–nucleic acid assemblies, and protein–ligand co-folding; use cases include target identification, antibody/protein engineering, and structure-based virtual screening.
- Runs on SageMaker Asynchronous Inference since folding often takes minutes: requests go to S3 via InvokeEndpointAsync with up to 1-hour timeout, results written to S3 with optional SNS notifications; internal queuing handles batches and scale-to-zero avoids idle costs.
- Defaults to ml.g6.4xlarge (2xlarge/12xlarge supported) with multi-GPU parallel predictions; you can supply your own MSAs for accuracy, and network isolation keeps sequences inside your account.
- Full walkthrough with code: ModelBuilder.fromjumpstartconfig, AsyncInferenceConfig output/failure paths, least-privilege IAM setup, quota checks, and cleanup guidance.
For computational drug discovery teams, this offers a private structure-prediction pipeline without managing GPU/CUDA/PyTorch stacks yourself.
More from Infra
- Redditor predicts Chinese AI chips will flood the market as Nvidia stumbles — Apart_Tumbleweed_303 · 2026-11-14
- FAA issues ground stop, likened to "Microsoft yesterday" systemwide outage — Kyrannio · 2026-11-14
- GAO: 76% of US air traffic control systems unsustainable, still on floppies and Win95 — MartinGTobias · 2026-11-14
- MoE combine traffic cut from 16 to 8.25 bits per value via RHT + S2F6 quantization — snowclipsed · 2026-11-14
- Report claims Jensen Huang resigns as Nvidia CEO; CFO Colette Kress named interim CEO — pdamodaran · 2026-11-14
- NVIDIA lands nvfp4 in large-scale pretraining under codename "yolo bundle" — SonglinYang4 · 2026-11-14