JapanFold offers 8 free open-source biology AI models on Japan-hosted inference infra
DavidBennett__ · x · 2026-09-03
ai& team's Shimpei Hara and David Bennett explain how they run eight open-source biology models free on hardware they own and operate in Japan, covering structure prediction, molecular binding, binder design, and protein search/classification, accessible via browser, API, or agents.
- AlphaFold turned protein structure prediction into research infrastructure with 200M+ public predicted structures; AI's role has since widened to binding prediction, protein design, and large-scale sequence search
- These capabilities are being connected into a single computational workflow, letting researchers triage candidates before bench work and focus limited time and budget on promising directions
- The piece details why serving biology models looks nothing like serving LLMs: a single structure prediction requires searching enormous sequence databases, computing amino-acid relationships, and generating multiple candidate structures
More from Infra
- Nvidia's vertical integration signals next-gen app layer push beyond CUDA — RachelVT42 · 2026-09-03
- ARK mid-year AI review: compute buildout and adoption keep beating expectations — downingARK · 2026-09-03
- Buying a refurbished 8xA100 server to colocate and rent out: one user's GPU math — Alarming-Ad8154 · 2026-09-03
- Hidden consumable under AI servers: PCB drill-bit demand is growing far faster than board volumes — tengyanAI · 2026-09-03
- Vera Rubin NVL72 hits up to 10X tokens per MW vs GB200 on DeepSeek R1, CoreWeave data shows — Beth_Kindig · 2026-09-03
- Tencent Hunyuan 770B compressed from ~1.5TB to ~214GiB with mixed quantization — Aiden_Tech_Ai · 2026-09-03