[1/n][CI] Load models in CI from S3 instead of HF (#13205)

Signed-off-by: <>
Co-authored-by: EC2 Default User <ec2-user@ip-172-31-20-117.us-west-2.compute.internal>

tests/basic_correctness/test_preemption.py

@@ -17,7 +17,7 @@ from vllm.core.scheduler import (ARTIFICIAL_PREEMPTION_MAX_CNT,
 from ..models.utils import check_outputs_equal
 
 MODELS = [
-    "facebook/opt-125m",
+    "distilbert/distilgpt2",
 ]