Improve weight loading and code style (#3174)

This commit is contained in:
Lianmin Zheng
2025-01-27 03:00:41 -08:00
committed by GitHub
parent 351a72d40b
commit 53cef81587
11 changed files with 171 additions and 65 deletions

View File

@@ -71,8 +71,8 @@ nvcc_flags = [
"-std=c++17",
"-use_fast_math",
"-DFLASHINFER_ENABLE_F16",
"-Xcompiler",
"-w",
"-Xcompiler=-Wconversion",
"-Xcompiler=-fno-strict-aliasing",
]
nvcc_flags_fp8 = [
"-DFLASHINFER_ENABLE_FP8",