fix(container): update image ghcr.io/ggml-org/llama.cpp (server-rocm-b10795 ➔ server-rocm-b10818)

[renovate]

This PR contains the following updates:

Package	Update	Change
ghcr.io/ggml-org/llama.cpp	patch	server-rocm-b10795 → server-rocm-b10818

---

Warning

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

---

Configuration

📅 Schedule: (in timezone Europe/Brussels)

• 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 this update again.

---

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

---

This PR was generated by Mend Renovate. View the repository job log.

[deepsource-io]

DeepSource Code Review

We reviewed changes in 139ddc2...2def0d7 on this pull request. Below is the summary for the review, and you can see the individual issues we found as inline review comments.

See full review on DeepSource ↗

PR Report Card

Overall Grade

Security   Reliability   Complexity   Hygiene

Code Review Summary

Analyzer	Status	Updated (UTC)	Details
JavaScript		Sep 5, 2026 5:32p.m.	Review ↗
Shell		Sep 5, 2026 5:32p.m.	Review ↗

---

Important

AI Review is run only on demand for your team. We're only showing results of static analysis review right now. To trigger AI Review, comment @deepsourcebot review on this thread.

[tanguille-cluster]

@@ spec.image @@
# inference.llmkube.dev/v1alpha1/InferenceService/ai/qwen38-flash-next
! ± value change
- ghcr.io/ggml-org/llama.cpp:server-rocm-b10795
+ ghcr.io/ggml-org/llama.cpp:server-rocm-b10818

[github-actions]

AI Automated Review

Analysis engine: omniroute@http://litellm.ai.svc.cluster.local/v1 (openai)

Approve upgrade of ghcr.io/ggml-org/llama.cpp from server-rocm-b10795 to server-rocm-b10818 in kubernetes/apps/ai/llmkube/models/qwen38-flash-next.yaml.

Change-by-change findings

• kubernetes/apps/ai/llmkube/models/qwen38-flash-next.yaml: image tag updated to server-rocm-b10818.

Sources

• PR PR 4877 diff and metadata
• Upstream ggml-org/llama.cpp releases and compare API (b10795...b10818)

Tool Harness Findings

• Executed read_file on model manifest and gh_api calls for upstream release comparison successfully.

Release notes

• Upstream commits between b10795 and b10818 provide ROCm backend optimizations, kernel fusion fixes, and stability improvements.