scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-04 02:31:27 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 69

fix(streaming): drop zero usage.cost in calculate_total_usage

Browse source 
Greptile P1: Vertex stream chunks can leave usage.cost=0 on assembled
usage; stamping that value bypasses token-based stream spend. Reuse
_resolve_provider_reported_cost so non-positive leftovers are omitted.
This commit is contained in:
leilei3167 2026-09-23 05:06:31 +00:00
parent fd855b2b57
commit 8567010897
1 changed files with 3 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
litellm/litellm_core_utils/streaming_handler.py
View file

@ -2604,8 +2604,9 @@ def calculate_total_usage(chunks: list[ModelResponse]) -> Usage:
if isinstance(latest_usage_chunk, dict)
else getattr(latest_usage_chunk, "cost", None)
)
if latest_cost is not None:
returned_usage_chunk.cost = latest_cost
resolved_cost: Final = CustomStreamWrapper._resolve_provider_reported_cost(latest_cost)
if resolved_cost is not None:
returned_usage_chunk.cost = resolved_cost
return returned_usage_chunk