Updates default CUDA architectures to include additional versions for improved compatibility

Author: algo-home

setup.py

@@ -79,7 +79,7 @@ def should_skip_cuda_build():
 
 @functools.lru_cache(maxsize=None)
 def cuda_archs():
-    return os.getenv("FLASH_DMATTN_CUDA_ARCHS", "80;90").split(";")
+    return os.getenv("FLASH_DMATTN_CUDA_ARCHS", "80;90;100;120").split(";")
 
 
 def detect_preferred_sm_arch() -> Optional[str]: