fast_trimul

fast_trimul

A drop-in library for fast Triangle Multiplicative Updates

4 followers

A drop-in, hardware-agnostic library for Fused Triangle Multiplicative Updates across AlphaFold3 family models, powered by CuTe DSL. Uses roughly half the GPU memory with almost zero activation data at any sequence length. It is far faster for small sequences.

fast_trimul

Launch date
fast_trimul
fast_trimulA drop-in library for fast Triangle Multiplicative Updates

Launched on August 19th, 2026