A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[CI/Build] Fix amd model executor test (#27612)
Signed-off-by: zhewenli <zhewenli@meta.com>
Zhewen Li committed
Oct 28, 2025 at 01:58 UTC
0291fbf65cec9082dba4358f33f1d89ef595172c
3 files changed
+12
-1
.buildkite/test-amd.yaml
+1
-1
@@ -561,7 +561,7 @@ steps:
561
562
- label: Model Executor Test # 23min
563
timeout_in_minutes: 35
564
- mirror_hardwares: [amdexperimental]
564
+ mirror_hardwares: [amdexperimental, amdproduction]
565
agent_pool: mi325_1
566
# grade: Blocking
567
source_file_dependencies:
tests/model_executor/model_loader/fastsafetensors_loader/test_fastsafetensors_loader.py
+6
@@ -1,7 +1,10 @@
1
# SPDX-License-Identifier: Apache-2.0
2
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
3
4
+import pytest
5
+
6
from vllm import SamplingParams
7
+from vllm.platforms import current_platform
8
9
test_model = "openai-community/gpt2"
10
@@ -15,6 +18,9 @@ prompts = [
18
sampling_params = SamplingParams(temperature=0.8, top_p=0.95, seed=0)
19
20
21
+@pytest.mark.skipif(
22
+ not current_platform.is_cuda(), reason="fastsafetensors requires CUDA/NVIDIA GPUs"
23
+)
24
def test_model_loader_download_files(vllm_runner):
25
with vllm_runner(test_model, load_format="fastsafetensors") as llm:
26
deserialized_outputs = llm.generate(prompts, sampling_params)
tests/model_executor/model_loader/fastsafetensors_loader/test_weight_utils.py
+5
@@ -5,6 +5,7 @@ import glob
5
import tempfile
6
7
import huggingface_hub.constants
8
+import pytest
9
import torch
10
11
from vllm.model_executor.model_loader.weight_utils import (
@@ -12,8 +13,12 @@ from vllm.model_executor.model_loader.weight_utils import (
13
fastsafetensors_weights_iterator,
14
safetensors_weights_iterator,
15
)
16
+from vllm.platforms import current_platform
17
18
19
+@pytest.mark.skipif(
20
+ not current_platform.is_cuda(), reason="fastsafetensors requires CUDA/NVIDIA GPUs"
21
+)
22
def test_fastsafetensors_model_loader():
23
with tempfile.TemporaryDirectory() as tmpdir:
24
huggingface_hub.constants.HF_HUB_OFFLINE = False