biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Frontend] Add Early Validation For Chat Template / Tool Call Parser (#9151)

Browse Source 
Signed-off-by: Alex-Brooks <Alex.Brooks@ibm.com>
This commit is contained in:
Alex Brooks
2024-10-08 08:31:26 -06:00
committed by GitHub
parent a3691b6b5e
commit 069d3bd8d0
5 changed files with 153 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
vllm/entrypoints/openai/api_server.py
View File

@@ -31,7 +31,8 @@ from vllm.engine.multiprocessing.engine import run_mp_engine
from vllm.engine.protocol import EngineClient
from vllm.entrypoints.launcher import serve_http
from vllm.entrypoints.logger import RequestLogger
from vllm.entrypoints.openai.cli_args import make_arg_parser
from vllm.entrypoints.openai.cli_args import (make_arg_parser,
validate_parsed_serve_args)
# yapf conflicts with isort for this block
# yapf: disable
from vllm.entrypoints.openai.protocol import (ChatCompletionRequest,
@@ -577,5 +578,6 @@ if __name__ == "__main__":
description="vLLM OpenAI-Compatible RESTful API server.")
parser = make_arg_parser(parser)
args = parser.parse_args()
validate_parsed_serve_args(args)
uvloop.run(run_server(args))