ViperEkura 0dd9a417b7 refactor: separate KV token address resolution

2026-08-15 22:59:35 +08:00

..

kernels

refactor: separate KV token address resolution

2026-08-15 22:59:35 +08:00

tests

perf: bypass L1 for attention tile loads

2026-08-15 21:23:20 +08:00

__init__.py

feat: add optional CUDA kernel system (csrc/) + fused GQA decode attention

2026-07-06 12:09:58 +08:00

CMakeLists.txt

feat: add fp8 training via cublasLt dispatch

2026-08-14 00:39:49 +08:00