Improve streaming, log_level, memory report, weight loading, and benchmark script (#7632)

Co-authored-by: Kan Wu <wukanustc@gmail.com>
This commit is contained in:
Lianmin Zheng
2025-06-29 23:16:19 -07:00
committed by GitHub
parent c5131f7a2f
commit 22352d47a9
24 changed files with 626 additions and 160 deletions

View File

@@ -516,9 +516,6 @@ class EmbeddingReqInput:
# For cross-encoder requests
is_cross_encoder_request: bool = False
def contains_mm_input(self) -> bool:
return has_valid_data(self.image_data) or has_valid_data(self.audio_data)
def normalize_batch_and_arguments(self):
# at least one of text, input_ids, or image should be provided
if self.text is None and self.input_ids is None and self.image_data is None:
@@ -572,6 +569,9 @@ class EmbeddingReqInput:
self.rid = uuid.uuid4().hex
return self.rid
def contains_mm_input(self) -> bool:
return has_valid_data(self.image_data) or has_valid_data(self.audio_data)
def __getitem__(self, i):
if self.is_cross_encoder_request:
return EmbeddingReqInput(