mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-27 06:13:07 +00:00
9026f42710
The team Model-Specific Rate Limits picker built its options from the
team's direct models only. A team that gets its models through access
groups has models ["no-default-models"], so the sentinel was the only
option, and a saved {"no-default-models": N} limit never enforces since
the limiter matches the requested model name.
Build one option list for every team per-model field from direct plus
access-group models, expand wildcards, treat an empty list as all
models, and drop sentinels and wildcard patterns, which no limit or
budget lookup ever matches. The team per-model budget field shared the
broken list and is fixed with it.
LIT-9490
|
||
|---|---|---|
| .. | ||
| docker-entrypoint.d | ||
| litellm-dashboard | ||
| Dockerfile | ||
| nginx.conf | ||