2025-07-25 –, Lawrence Room 107 - Method Room
This talk introduces Reactant, a tool for optimizing existing Julia programs through MLIR and XLA. Reactant preserves high level semantics already found in Julia programs (like linear algebra) to enable novel optimizations that cross function boundaries, rewrite programs to effectively execute on any hardware backend (including rewriting CUDA programs to run on thousands of distributed TPUs), and combines with Enzyme to provide high-performance multi-backend differentiation.
This talk introduces Reactant, a tool for optimizing existing Julia programs through MLIR and XLA. Reactant preserves high level semantics already found in Julia programs (like linear algebra) to enable novel optimizations that cross function boundaries, rewrite programs to effectively execute on any hardware backend (including rewriting CUDA programs to run on thousands of distributed TPUs), and combines with Enzyme to provide high-performance multi-backend differentiation.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.