Terray taps NVIDIA CUDA kernels to speed drug-design model TerraBind up to 7x
AllThingsApx · x · 2026-11-17
Biotech company Terray details its collaboration with NVIDIA to upgrade TerraBind, the universal potency model behind its EMMI drug-design platform.
- Integrating NVIDIA CuTeDSL kernels in production cut memory overhead, eliminated out-of-memory bottlenecks on massive target complexes, and accelerated inference by up to 7x
- Using the new cuPotential library for molecular dynamics, synthetic training data generation got roughly 50% more efficient, enabling adaptive in-silico data generation on the fly
- Potency modeling historically forced a speed-vs-accuracy trade-off in chemical space exploration; this upgrade pushes both fronts simultaneously
More from Infra
- tinygrad.js runs the real tinygrad in your browser on your GPU via WebGPU — CatAstro_Piyush · 2026-11-17
- Data Center Backlash Grows Loud, With Historical Echoes of 'A Nuisance Has No Rights' — justin_hart · 2026-11-17
- Constrained Decoding for Function Calling Cuts Endpoint Throughput by 17%-90% — remilouf · 2026-11-17
- Big Tech locks up 60% of DRAM supply it can't deploy yet — Reddit proposal for use-it-or-lose-it auctions — politefella0 · 2026-11-17
- NVIDIA expands CUDA-X science libraries: black hole analysis drops from a month to 8 minutes — AllThingsApx · 2026-11-17
- GPUs resell in weeks; the real AI credit risk is half-built campuses with debt and no tenants — tengyanAI · 2026-11-17