mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-11 03:38:38 +00:00
Replace per-member inline prisma loops on PATCH /v2/team/{team_id}/members
with a pure write planner and set-oriented apply path. Load memberships and
budgets once, plan disconnect/update/create-and-attach ops without prisma,
then apply in one transaction via create_many plus attach/update. Also
invalidate team membership caches after bulk updates so auth does not keep
stale allowed_models or limits
|
||
|---|---|---|
| .. | ||
| management | ||
| __init__.py | ||