litellm/enterprise/litellm_enterprise/proxy/common_utils
joshua-berri 3fb95c8fe3
fix(batches): preserve spend across malformed output and metadata failures (#47184)
* fix(batches): preserve recoverable spend and terminal reconciliation state

* fix(batches): narrow validated envelope types

* fix(batches): serialize stored metadata when releasing claims

* fix(batches): sanitize metadata before claiming completed batches

* refactor(batches): expose existing metadata serializer for shared use

---------

Co-authored-by: Joshua Valluru <326636767+joshua-berri@users.noreply.github.com>
2026-10-17 17:10:22 -07:00
..
__init__.py Normalize OpenAI SDK BaseModel choices/messages to avoid Pydantic serializer warnings (#18972) 2026-01-14 03:40:11 +05:30
check_batch_cost.py fix(batches): preserve spend across malformed output and metadata failures (#47184) 2026-10-17 17:10:22 -07:00
check_responses_cost.py fix(responses): bill a background response once, on the first read or poll that sees it finished (#45703) 2026-10-15 11:00:14 -07:00
managed_object_attribution.py fix(responses): bill a background response once, on the first read or poll that sees it finished (#45703) 2026-10-15 11:00:14 -07:00