fix: preserve priority for batched embedding requests (#32977)
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 5
parent
db8f3cdd11
commit
fe6a05a8e8
@@ -1161,6 +1161,7 @@ class EmbeddingReqInput:
|
||||
lora_id=self.lora_id[i] if self.lora_id is not None else None,
|
||||
positional_embed_overrides=self._get_positional_embed_overrides_item(i),
|
||||
http_worker_ipc=self.http_worker_ipc,
|
||||
priority=self.priority,
|
||||
return_pooled_hidden_states=self.return_pooled_hidden_states,
|
||||
return_prompt_token_ids=self.return_prompt_token_ids,
|
||||
multi_item_delimiter_indices=(
|
||||
@@ -1188,6 +1189,7 @@ class EmbeddingReqInput:
|
||||
lora_id=self.lora_id[i] if self.lora_id is not None else None,
|
||||
positional_embed_overrides=self._get_positional_embed_overrides_item(i),
|
||||
http_worker_ipc=self.http_worker_ipc,
|
||||
priority=self.priority,
|
||||
dimensions=self.dimensions,
|
||||
return_pooled_hidden_states=self.return_pooled_hidden_states,
|
||||
return_prompt_token_ids=self.return_prompt_token_ids,
|
||||
|
||||
Reference in New Issue
Block a user