scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-10 22:41:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 83
litellm/tests/test_litellm/completion_extras
History
Exact
mateo-berri 3448175184 fix(responses bridge): keep mid-conversation system messages in input 
Only the leading run of system messages (before the first non-system
message) is joined into the Responses `instructions` field. A system
message that arrives after a user, assistant, or tool turn now becomes a
system input item at its position, whether its content is a string or a
list, so a client that re-sends the same reminder as a string on the
next request produces byte-identical input and `instructions` stays
stable. Claude Code >= 2.1.237 appends such reminders after every user
turn, and folding them into `instructions` made Azure treat every
request as a cold prompt (cached_tokens 0 on every request of a
session).

Fixes #40198
2026-09-08 11:44:49 -07:00
..
litellm_responses_transformation fix(responses bridge): keep mid-conversation system messages in input 2026-09-08 11:44:49 -07:00
__init__.py Litellm oss staging 030626 (#29578) 2026-06-03 11:01:51 -07:00
test_litellm_responses_transformation_transformation.py fix(litellm): support OpenAI chat completions custom tool calls end to end 2026-08-01 11:26:09 -07:00
test_responses_bridge_provider_propagation.py fix: bedrock mantle fixes (#30083) 2026-06-12 07:25:16 -07:00