Lists (1)
Stars
2
stars
written in Cuda
Clear filter
🤖FFPA: Extend FlashAttention-2 with Split-D, ~O(1) SRAM complexity for large headdim, 1.8x~3x↑🎉 vs SDPA EA.
Parrot is a C++ library for fused array operations using CUDA/Thrust. It provides efficient GPU-accelerated operations with lazy evaluation semantics, allowing for chaining of operations without un…