[Frontend] Consolidate scale out entrypoints (#44512)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
@@ -3,7 +3,7 @@
|
||||
"""Unit tests for the token-offsets request/response protocol wiring:
|
||||
the request flag flowing into ``TokenizeParams`` and the ``GenerateRequest``
|
||||
serialization boundary. End-to-end behavior is covered by
|
||||
``tests/entrypoints/serve/render/test_render.py``; plain Pydantic field
|
||||
``tests/entrypoints/scale_out/render/test_render.py``; plain Pydantic field
|
||||
storage is not retested here.
|
||||
"""
|
||||
|
||||
@@ -12,7 +12,7 @@ from unittest.mock import Mock
|
||||
from vllm.config import ModelConfig
|
||||
from vllm.entrypoints.openai.chat_completion.protocol import ChatCompletionRequest
|
||||
from vllm.entrypoints.openai.completion.protocol import CompletionRequest
|
||||
from vllm.entrypoints.serve.disagg.protocol import GenerateRequest
|
||||
from vllm.entrypoints.scale_out.token_in_token_out.protocol import GenerateRequest
|
||||
from vllm.sampling_params import SamplingParams
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user