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

[Misc] Split up pooling tasks (#10820)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2024-12-11 17:28:00 +08:00
committed by GitHub
parent 40766ca1b8
commit 8f10d5e393
27 changed files with 527 additions and 168 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
vllm/entrypoints/openai/api_server.py
View File

@@ -573,7 +573,7 @@ def init_app_state(
enable_auto_tools=args.enable_auto_tool_choice,
tool_parser=args.tool_call_parser,
enable_prompt_tokens_details=args.enable_prompt_tokens_details,
) if model_config.task == "generate" else None
) if model_config.runner_type == "generate" else None
state.openai_serving_completion = OpenAIServingCompletion(
engine_client,
model_config,
@@ -582,7 +582,7 @@ def init_app_state(
prompt_adapters=args.prompt_adapters,
request_logger=request_logger,
return_tokens_as_token_ids=args.return_tokens_as_token_ids,
) if model_config.task == "generate" else None
) if model_config.runner_type == "generate" else None
state.openai_serving_embedding = OpenAIServingEmbedding(
engine_client,
model_config,
@@ -590,13 +590,13 @@ def init_app_state(
request_logger=request_logger,
chat_template=resolved_chat_template,
chat_template_content_format=args.chat_template_content_format,
) if model_config.task == "embedding" else None
) if model_config.runner_type == "pooling" else None
state.openai_serving_scores = OpenAIServingScores(
engine_client,
model_config,
base_model_paths,
request_logger=request_logger
) if (model_config.task == "embedding" \
) if (model_config.runner_type == "pooling" \
and model_config.is_cross_encoder) else None
state.openai_serving_tokenization = OpenAIServingTokenization(
engine_client,