feat(container): update image ghcr.io/berriai/litellm-non_root (v1.95.0 ➔ v1.96.0)

[drag0n141-bot]

This PR contains the following updates:

Package	Update	Change
ghcr.io/berriai/litellm-non_root (source)	minor	v1.95.0 → v1.96.0

---

Warning

Some dependencies could not be looked up. Check the Dependency Dashboard for more information.

---

Configuration

📅 Schedule: (in timezone Europe/Berlin)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about these updates again.

---

• If you want to rebase/retry this PR, check this box

---

This PR has been generated by Mend Renovate CLI.

[drag0n141-bot]

--- kubernetes/apps/ai/litellm/app Kustomization: ai/litellm HelmRelease: ai/litellm

+++ kubernetes/apps/ai/litellm/app Kustomization: ai/litellm HelmRelease: ai/litellm

@@ -42,13 +42,13 @@

               TZ: Europe/Berlin
             envFrom:
             - secretRef:
                 name: litellm-secret
             image:
               repository: ghcr.io/berriai/litellm-non_root
-              tag: v1.95.0@sha256:d5233c80ef7e28fd789b93c15e41fc0c86c6dfdf6ecbe1ef3f4c8ee2498f3827
+              tag: v1.96.0@sha256:8b64d0fa562739649932b6f654a03dee9a511e7dd0a60bdba9d37b8fd3ffd69d
             probes:
               liveness:
                 custom: true
                 enabled: true
                 spec:
                   failureThreshold: 3

[drag0n141-bot]

--- HelmRelease: ai/litellm Deployment: ai/litellm

+++ HelmRelease: ai/litellm Deployment: ai/litellm

@@ -55,13 +55,13 @@

           value: '6379'
         - name: TZ
           value: Europe/Berlin
         envFrom:
         - secretRef:
             name: litellm-secret
-        image: ghcr.io/berriai/litellm-non_root:v1.95.0@sha256:d5233c80ef7e28fd789b93c15e41fc0c86c6dfdf6ecbe1ef3f4c8ee2498f3827
+        image: ghcr.io/berriai/litellm-non_root:v1.96.0@sha256:8b64d0fa562739649932b6f654a03dee9a511e7dd0a60bdba9d37b8fd3ffd69d
         livenessProbe:
           failureThreshold: 3
           httpGet:
             path: /
             port: 4000
           initialDelaySeconds: 5

[claude]

litellm-non_root: v1.95.0 → v1.96.0

Verdict: Safe to merge

Minor version bump of the BerriAI LiteLLM proxy image. No formal GitHub Release exists yet for the final v1.96.0 tag (only v1.96.0-rc.1/dev.* pre-releases, which the image tag appears to match — the rc.1...dev.2 diff is empty, so rc.1's changelog is the accurate content for this build).

Reviewed the full commit list between the two versions. One breaking change was flagged (feat(proxy)!: gate all mock testing request params behind a single config flag, #35423): the mock_timeout and mock_testing_rate_limit_error request body params, previously accepted unconditionally, now return HTTP 400 unless general_settings.dangerously_allow_mock_testing_request_params: true is set. This repo's kubernetes/apps/ai/litellm/app/resources/config.yaml does not set that flag and none of the configured clients appear to send mock-testing params, so this does not affect normal traffic — it only blocks a testing feature this deployment doesn't use.

Other changes in this range (componentized Helm chart hardening/probe fixes, config-sync-via-redis-pubsub, aiohttp floor bump, prisma/DB fixes) are either specific to the official LiteLLM Helm chart (this deployment uses app-template instead) or internal implementation details with no config surface this repo touches.

New features worth adopting (optional, not required for this upgrade):

• feat(prometheus): add global exclude_metrics and exclude_labels options (#34201) — this repo enables success_callback: ["prometheus"] and prometheus_adapter: true; could be used later to trim metric cardinality if needed. Would touch kubernetes/apps/ai/litellm/app/resources/config.yaml.
• feat(proxy): push config sync to pods via redis pub/sub (#35436) — only relevant if this deployment ever runs multiple replicas (currently single replica), since Redis is already deployed alongside LiteLLM.

Sources consulted:

• https://github.com/BerriAI/litellm/releases/tag/v1.96.0-rc.1
• BerriAI/litellm#35423
• BerriAI/litellm@v1.96.0-dev.2...v1.96.0-rc.1