mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2026-03-16 11:47:09 +08:00
[BugFix] [FEAT] Enable fastsafetensors for ROCm platform (#28225)
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
This commit is contained in:
parent
a2e9ebe9e2
commit
82b05b15e6
@ -15,3 +15,4 @@ setuptools-scm>=8
|
|||||||
runai-model-streamer[s3,gcs]==0.15.0
|
runai-model-streamer[s3,gcs]==0.15.0
|
||||||
conch-triton-kernels==1.2.1
|
conch-triton-kernels==1.2.1
|
||||||
timm>=1.0.17
|
timm>=1.0.17
|
||||||
|
fastsafetensors @ git+https://github.com/foundation-model-stack/fastsafetensors.git@d6f998a03432b2452f8de2bb5cefb5af9795d459
|
||||||
|
|||||||
@ -19,7 +19,8 @@ sampling_params = SamplingParams(temperature=0.8, top_p=0.95, seed=0)
|
|||||||
|
|
||||||
|
|
||||||
@pytest.mark.skipif(
|
@pytest.mark.skipif(
|
||||||
not current_platform.is_cuda(), reason="fastsafetensors requires CUDA/NVIDIA GPUs"
|
not current_platform.is_cuda_alike(),
|
||||||
|
reason="fastsafetensors requires NVIDIA/AMD GPUs",
|
||||||
)
|
)
|
||||||
def test_model_loader_download_files(vllm_runner):
|
def test_model_loader_download_files(vllm_runner):
|
||||||
with vllm_runner(test_model, load_format="fastsafetensors") as llm:
|
with vllm_runner(test_model, load_format="fastsafetensors") as llm:
|
||||||
|
|||||||
@ -17,7 +17,8 @@ from vllm.platforms import current_platform
|
|||||||
|
|
||||||
|
|
||||||
@pytest.mark.skipif(
|
@pytest.mark.skipif(
|
||||||
not current_platform.is_cuda(), reason="fastsafetensors requires CUDA/NVIDIA GPUs"
|
not current_platform.is_cuda_alike(),
|
||||||
|
reason="fastsafetensors requires NVIDIA/AMD GPUs",
|
||||||
)
|
)
|
||||||
def test_fastsafetensors_model_loader():
|
def test_fastsafetensors_model_loader():
|
||||||
with tempfile.TemporaryDirectory() as tmpdir:
|
with tempfile.TemporaryDirectory() as tmpdir:
|
||||||
|
|||||||
Loading…
x
Reference in New Issue
Block a user