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 15, 13:30 UTC
Fix OOT registration test (#17099)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Harry Mellor committed
Apr 24, 2025 at 12:44 UTC
a9138e85b14047e06300685b48e3485b995425fb
1 file changed
+2
-3
tests/models/test_oot_registration.py
+2
-3
@@ -18,10 +18,9 @@ def test_plugin(
18
m.setenv("VLLM_USE_V1", "0")
19
m.setenv("VLLM_PLUGINS", "")
20
21
- with pytest.raises(Exception) as excinfo:
21
+ match = "Cannot find model module"
22
+ with pytest.raises(ValueError, match=match):
23
LLM(model=dummy_opt_path, load_format="dummy")
23
- error_msg = "has no vLLM implementation and the Transformers implementation is not compatible with vLLM" # noqa: E501
24
- assert (error_msg in str(excinfo.value))
24
25
26
@create_new_process_for_each_test()