[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/mq_llm_engine/test_error_handling.py

@@ -21,8 +21,10 @@ from vllm.lora.request import LoRARequest
 from vllm.usage.usage_lib import UsageContext
 from vllm.utils import FlexibleArgumentParser
 
-MODEL = "google/gemma-1.1-2b-it"
-ENGINE_ARGS = AsyncEngineArgs(model=MODEL, enforce_eager=True)
+MODEL = "s3://vllm-ci-model-weights/gemma-1.1-2b-it"
+ENGINE_ARGS = AsyncEngineArgs(model=MODEL,
+                              load_format="runai_streamer",
+                              enforce_eager=True)
 RAISED_ERROR = KeyError
 RAISED_VALUE = "foo"