Tenstorrent runs Proteina protein model at high throughput, claims no HBM shortage
gajesh · x · 2026-10-27
RFD4-Proteina, the Baker Lab's open-source protein generation model, now runs on Tenstorrent hardware via TT-Bio, live on Japanfold at high throughput and low cost.
- The Baker Lab ported it to Tenstorrent on day 0 after release
- Tenstorrent claims highest throughput at far lower cost, with no exposure to the HBM shortage and near-zero wait times at any scale
- Supporters predict Tenstorrent could become the default hardware for bio
A notable real-world deployment of Tenstorrent as an alternative to NVIDIA's ecosystem.
More from Infra
- Sail equivalence proof and riscv-coq cross-checks uncover real RISC-V spec bugs — xennygrimmato_ · 2026-10-27
- xAI's Colossus 1 hits ~14.6 exaflops FP64, ~5x Top500 leader — cis_female · 2026-10-27
- Lambda runs 19 B200 nodes in a 16-node power budget, boosting throughput 24% — TheZachMueller · 2026-10-27
- Podcast: Micron's HBM roadmap, Meta's AI infra, and NVIDIA's supply chain lock-in — BenBajarin · 2026-10-27
- SGLang MegaKernel + PD Pushes the TCO-Latency Frontier on GB300, SemiAnalysis Benchmarks Show — AccBalanced · 2026-10-27
- Scaling self-hosted Convex: compiling TypeScript Convex functions to WASM — leventov · 2026-10-27