Research · BAIR Berkeley ·

From CUDA to MLX: How K-Search Brings Decades of Kernel Expertise to Apple Silicon

From CUDA to MLX: How K-Search Brings Decades of Kernel Expertise to Apple Silicon

BAIR researchers extend K-Search, an evolutionary kernel optimization framework, to translate CUDA optimization knowledge into architecture-native MLX strategies for Apple Silicon. The work explores automating the transfer of GPU kernel expertise across hardware ecosystems.

Read the full story at BAIR Berkeley →