[Bugfix] complete VLLMValidationError migration in chat_utils.py (#50254)
Signed-off-by: shaolila <shaolila@buaa.edu.cn> Co-authored-by: shaolila <shaolila@buaa.edu.cn>
This commit is contained in:
@@ -11,6 +11,7 @@ from vllm.entrypoints.launchers.cli_args import (
|
||||
validate_parsed_serve_args,
|
||||
)
|
||||
from vllm.entrypoints.openai.models.protocol import LoRAModulePath
|
||||
from vllm.exceptions import VLLMValidationError
|
||||
from vllm.utils.argparse_utils import FlexibleArgumentParser
|
||||
|
||||
LORA_MODULE = {
|
||||
@@ -204,7 +205,7 @@ def test_chat_template_validation_for_happy_paths(serve_parser):
|
||||
def test_chat_template_validation_for_sad_paths(serve_parser):
|
||||
"""Ensure validation fails if the chat template doesn't exist"""
|
||||
args = serve_parser.parse_args(args=["--chat-template", "does/not/exist"])
|
||||
with pytest.raises(ValueError):
|
||||
with pytest.raises(VLLMValidationError):
|
||||
validate_parsed_serve_args(args)
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user