diff --git a/kubernetes/ai/llmkube/models/models.yaml b/kubernetes/ai/llmkube/models/models.yaml
index 1652857c4..13484c072 100644
--- a/kubernetes/ai/llmkube/models/models.yaml
+++ b/kubernetes/ai/llmkube/models/models.yaml
@@ -26,7 +26,7 @@ spec:
   runtime: llamacpp
   mode: chat
   replicas: 1
-  image: ghcr.io/ggml-org/llama.cpp:server-cuda@sha256:190d82d2f35ff25539f181328bae0659a21cbe4e321259193216ed7d2d7b5e67
+  image: ghcr.io/ggml-org/llama.cpp:server-cuda@sha256:a1c6b2c951e24c704a0d753286742f7b861cdd718e3dc8d7f70d259e1d57d935
   bindAddress: 0.0.0.0
   contextSize: 131072
   parallelSlots: 1