Update renovate/renovate Docker tag to v41.51.0

Reviewed-on: #63

apps/immich/release.yaml

@@ -18,7 +18,7 @@ spec:
   chart:
     spec:
       chart: immich
-      version: 0.7.0
+      version: 0.7.1
       sourceRef:
         kind: HelmRepository
         name: secustor
@@ -58,6 +58,7 @@ spec:
       annotations:
         cert-manager.io/cluster-issuer: letsencrypt
         nginx.org/client-max-body-size: "0"
+        nginx.org/websocket-services: immich-server
       hosts:
         - host: immich.lumpiasty.xyz
           paths:

apps/immich/volume.yaml

@@ -9,5 +9,5 @@ spec:
     - ReadWriteOnce
   resources:
     requests:
-      storage: 50Gi
+      storage: 150Gi
   storageClassName: mayastor-single-hdd

apps/librechat/release.yaml

@@ -2,11 +2,11 @@
 apiVersion: source.toolkit.fluxcd.io/v1
 kind: HelmRepository
 metadata:
-  name: bat-librechat
+  name: dynomite567-charts
   namespace: librechat
 spec:
   interval: 24h
-  url: https://charts.blue-atlas.de
+  url: https://dynomite567.github.io/helm-charts/
 ---
 apiVersion: helm.toolkit.fluxcd.io/v2
 kind: HelmRelease
@@ -18,10 +18,10 @@ spec:
   chart:
     spec:
       chart: librechat
-      version: 1.8.10
+      version: 1.8.9
       sourceRef:
         kind: HelmRepository
-        name: bat-librechat
+        name: dynomite567-charts
   values:
     global:
       librechat:
@@ -61,7 +61,7 @@ spec:
               modelDisplayLabel: "Ollama"
             - name: "Llama.cpp"
               apiKey: "llama"
-              baseURL: "http://llama.llama.svc.cluster.local:11434/v1/chat/completions"
+              baseURL: "http://llama.llama.svc.cluster.local:11434/v1"
               models:
                 default: [
                   "DeepSeek-R1-0528-Qwen3-8B-GGUF",
@@ -69,7 +69,10 @@ spec:
                   "Qwen3-8B-GGUF-no-thinking",
                   "gemma3n-e4b",
                   "gemma3-12b",
-                  "gemma3-12b-novision"
+                  "gemma3-12b-q2",
+                  "gemma3-12b-novision",
+                  "gemma3-4b",
+                  "gemma3-4b-novision"
                 ]
               titleConvo: true
               titleModel: "current_model"
@@ -87,6 +90,9 @@ spec:
       className: nginx
       annotations:
         cert-manager.io/cluster-issuer: letsencrypt
+        nginx.org/client-max-body-size: "0"
+        nginx.ingress.kubernetes.io/proxy-buffering: "false"
+        nginx.org/proxy-read-timeout: 30m
       hosts:
         - host: librechat.lumpiasty.xyz
           paths:

apps/llama/configs/config.yaml

@@ -1,24 +1,33 @@
+healthCheckTimeout: 600
+
 models:
   "DeepSeek-R1-0528-Qwen3-8B-GGUF":
+    ttl: 600
     cmd: |
       /app/llama-server
       -hf unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF:Q4_K_M
       -ngl 37 -c 16384
+      --no-warmup
       --port ${PORT}
   "Qwen3-8B-GGUF":
+    ttl: 600
     cmd: |
       /app/llama-server
       -hf unsloth/Qwen3-8B-GGUF:Q4_K_M
       -ngl 37 -c 16384
+      --no-warmup
       --port ${PORT}
   "Qwen3-8B-GGUF-no-thinking":
+    ttl: 600
     cmd: |
       /app/llama-server
       -hf unsloth/Qwen3-8B-GGUF:Q4_K_M
       -ngl 37 -c 16384
       --jinja --chat-template-file /config/qwen_nothink_chat_template.jinja
+      --no-warmup
       --port ${PORT}
   "gemma3n-e3b":
+    ttl: 600
     cmd: |
       /app/llama-server
       -hf unsloth/gemma-3n-E4B-it-GGUF:UD-Q4_K_XL
@@ -31,11 +40,13 @@ models:
       --min-p 0.00
       --top-k 64
       --top-p 0.95
+      --no-warmup
       --port ${PORT}
   "gemma3-12b":
+    ttl: 600
     cmd: |
       /app/llama-server
-      -hf unsloth/gemma-3-12b-it-GGUF:Q3_K_M
+      -hf unsloth/gemma-3-12b-it-GGUF:Q4_K_M
       --ctx-size 16384
       --n-gpu-layers 99
       --prio 2
@@ -44,11 +55,13 @@ models:
       --min-p 0.00
       --top-k 64
       --top-p 0.95
+      --no-warmup
       --port ${PORT}
   "gemma3-12b-novision":
+    ttl: 600
     cmd: |
       /app/llama-server
-      -hf unsloth/gemma-3-12b-it-GGUF:Q3_K_M
+      -hf unsloth/gemma-3-12b-it-GGUF:Q4_K_M
       --ctx-size 16384
       --n-gpu-layers 99
       --prio 2
@@ -58,4 +71,51 @@ models:
       --top-k 64
       --top-p 0.95
       --no-mmproj
+      --no-warmup
+      --port ${PORT}
+  "gemma3-12b-q2":
+    ttl: 600
+    cmd: |
+      /app/llama-server
+      -hf unsloth/gemma-3-12b-it-GGUF:Q2_K_L
+      --ctx-size 16384
+      --n-gpu-layers 99
+      --prio 2
+      --temp 1.0
+      --repeat-penalty 1.0
+      --min-p 0.00
+      --top-k 64
+      --top-p 0.95
+      --no-warmup
+      --port ${PORT}
+  "gemma3-4b":
+    ttl: 600
+    cmd: |
+      /app/llama-server
+      -hf unsloth/gemma-3-4b-it-GGUF:Q4_K_M
+      --ctx-size 16384
+      --n-gpu-layers 99
+      --prio 2
+      --temp 1.0
+      --repeat-penalty 1.0
+      --min-p 0.00
+      --top-k 64
+      --top-p 0.95
+      --no-warmup
+      --port ${PORT}
+  "gemma3-4b-novision":
+    ttl: 600
+    cmd: |
+      /app/llama-server
+      -hf unsloth/gemma-3-4b-it-GGUF:Q4_K_M
+      --ctx-size 16384
+      --n-gpu-layers 99
+      --prio 2
+      --temp 1.0
+      --repeat-penalty 1.0
+      --min-p 0.00
+      --top-k 64
+      --top-p 0.95
+      --no-mmproj
+      --no-warmup
       --port ${PORT}

apps/ollama/release.yaml

@@ -18,7 +18,7 @@ spec:
   chart:
     spec:
       chart: ollama
-      version: 1.24.0
+      version: 1.25.0
       sourceRef:
         kind: HelmRepository
         name: ollama-helm

apps/renovate/cronjob.yaml

@@ -15,7 +15,7 @@ spec:
             - name: renovate
               # Update this to the latest available and then enable Renovate on
               # the manifest
-              image: renovate/renovate:41.43.7-full
+              image: renovate/renovate:41.51.0-full
               envFrom:
                 - secretRef:
                     name: renovate-gitea-token