scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-05 02:41:56 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 49

fix(streaming): clear stale response_cost when stream assemble defers

Browse source
This commit is contained in:
leilei3167 2026-09-16 15:27:29 +00:00
parent def3efa72b
commit fd855b2b57
1 changed files with 5 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

7
litellm/main.py
View file

@ -8739,9 +8739,12 @@ def _stream_builder_model_map_cost(response: ModelResponse) -> float | None:
def _set_stream_builder_response_cost(response: ModelResponse, logging_obj: Optional["Logging"]) -> None:
response_cost: Final = _stream_builder_response_cost(response, logging_obj)
if response_cost is None:
return
hidden_params: Final = response._hidden_params # pyright: ignore[reportPrivateUsage] # no public accessor
if response_cost is None:
# Non-positive usage.cost (or a logging_obj deferral) must leave the field
# absent so token-based pricing can run — clear any earlier stamp.
hidden_params.pop("response_cost", None)
return
hidden_params["response_cost"] = response_cost