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

[Bugfix][Refactor] Unify model management in frontend (#11660)

Browse Source 
Signed-off-by: Joe Runde <Joseph.Runde@ibm.com>
This commit is contained in:
Joe Runde
2024-12-31 18:21:51 -08:00
committed by GitHub
parent 0c6f998554
commit 4db72e57f6
15 changed files with 365 additions and 307 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
vllm/entrypoints/openai/serving_tokenization.py
View File

@@ -15,9 +15,8 @@ from vllm.entrypoints.openai.protocol import (DetokenizeRequest,
TokenizeRequest,
TokenizeResponse)
# yapf: enable
from vllm.entrypoints.openai.serving_engine import (BaseModelPath,
LoRAModulePath,
OpenAIServing)
from vllm.entrypoints.openai.serving_engine import OpenAIServing
from vllm.entrypoints.openai.serving_models import OpenAIServingModels
from vllm.logger import init_logger
logger = init_logger(__name__)
@@ -29,18 +28,15 @@ class OpenAIServingTokenization(OpenAIServing):
self,
engine_client: EngineClient,
model_config: ModelConfig,
base_model_paths: List[BaseModelPath],
models: OpenAIServingModels,
*,
lora_modules: Optional[List[LoRAModulePath]],
request_logger: Optional[RequestLogger],
chat_template: Optional[str],
chat_template_content_format: ChatTemplateContentFormatOption,
) -> None:
super().__init__(engine_client=engine_client,
model_config=model_config,
base_model_paths=base_model_paths,
lora_modules=lora_modules,
prompt_adapters=None,
models=models,
request_logger=request_logger)
self.chat_template = chat_template