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 12, 06:00 UTC
[Misc] Update conftest for entrypoints/sagemaker test folder (#29799)
Signed-off-by: Zuyi Zhao <zhaozuy@amazon.com>
Zuyi Zhao committed
Dec 1, 2025 at 19:56 UTC
53bf71b0f0e0fc57ec5879ea3d13cf93a024667a
1 file changed
+4
-1
tests/entrypoints/sagemaker/conftest.py
+4
-1
@@ -45,7 +45,10 @@ def basic_server_with_lora(smollm2_lora_files):
45
"64",
46
]
47
48
- envs = {"VLLM_ALLOW_RUNTIME_LORA_UPDATING": "True"}
48
+ envs = {
49
+ "VLLM_ALLOW_RUNTIME_LORA_UPDATING": "True",
50
+ "SAGEMAKER_ENABLE_STATEFUL_SESSIONS": "True",
51
+ }
52
with RemoteOpenAIServer(MODEL_NAME_SMOLLM, args, env_dict=envs) as remote_server:
53
yield remote_server
54