[Test][Quantization] Gate CUTLASS W4A8 kernel tests to SM90 (#58878)
Signed-off-by: Kevin Li <kevinbean99@gmail.com> Co-authored-by: Misha Goin <mgoin64@gmail.com>
This commit is contained in:
@@ -17,9 +17,8 @@ from vllm.platforms import current_platform
|
||||
from vllm.scalar_type import ScalarType, scalar_types
|
||||
from vllm.utils.torch_utils import set_random_seed
|
||||
|
||||
capability = current_platform.get_device_capability()
|
||||
IS_SUPPORTED_BY_GPU = (
|
||||
current_platform.is_cuda() and capability is not None and capability[0] >= 9
|
||||
current_platform.is_cuda() and current_platform.is_device_capability(90)
|
||||
)
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user