[Hardware][AMD][CI][Bugfix] Fix AMD Quantization test group (#31713)
Signed-off-by: Matthew Wong <Matthew.Wong2@amd.com>
This commit is contained in:
@@ -644,6 +644,9 @@ def test_compressed_tensors_2of4_sparse_compressed(vllm_runner, args_2of4):
|
||||
assert output
|
||||
|
||||
|
||||
@pytest.mark.skipif(
|
||||
not current_platform.is_cuda(), reason="This test is skipped on non-CUDA platform."
|
||||
)
|
||||
@pytest.mark.parametrize(
|
||||
"args",
|
||||
[
|
||||
@@ -762,7 +765,10 @@ def test_compressed_tensors_fp8_block_enabled(vllm_runner):
|
||||
|
||||
input_quant_op = qkv_proj.scheme.w8a8_block_fp8_linear.input_quant_op
|
||||
assert isinstance(input_quant_op, QuantFP8)
|
||||
assert input_quant_op._forward_method == input_quant_op.forward_cuda
|
||||
assert input_quant_op._forward_method in (
|
||||
input_quant_op.forward_cuda,
|
||||
input_quant_op.forward_hip,
|
||||
)
|
||||
|
||||
llm.apply_model(check_model)
|
||||
|
||||
|
||||
Reference in New Issue
Block a user