ai: fix QDRANT_URI env var, bump qdrant to v1.17.0 to match client

ai/docker-compose.yaml

@@ -1,10 +1,13 @@
 networks:
   ai-net:
     driver: bridge
+  ai-services:
+    name: ai-services
   ix-databases_shared-databases:
     external: true
   pangolin:
     external: true
+
 services:
   ollama:
     container_name: ollama
@@ -12,6 +15,7 @@ services:
     restart: unless-stopped
     networks:
       - ai-net
+      - ai-services
     ports:
       - "11434:11434"
     environment:
@@ -27,12 +31,14 @@ services:
             - driver: nvidia
               count: all
               capabilities: [gpu]
+
   openwebui:
     container_name: openwebui
     image: ghcr.io/open-webui/open-webui:main
     restart: unless-stopped
     networks:
       - ai-net
+      - ai-services
       - ix-databases_shared-databases
       - pangolin
     ports:
@@ -44,6 +50,8 @@ services:
       DATABASE_URL: postgresql://openwebui:${OPENWEBUI_DB_PASS}@shared-postgres:5432/openwebui
       ENABLE_SIGNUP: "true"
       ENABLE_LOGIN_FORM: "true"
+      VECTOR_DB: qdrant
+      QDRANT_URI: http://qdrant:6333
     volumes:
       - /mnt/docker-ssd/docker/appdata/openwebui:/app/backend/data
     depends_on:
@@ -54,33 +62,78 @@ services:
       timeout: 10s
       retries: 3
       start_period: 60s
-#   scriberr:
+
-#     container_name: scriberr
+  qdrant:
-#     image: ghcr.io/rishikanthc/scriberr-cuda:v1.2.0
+    container_name: qdrant
-#     restart: unless-stopped
+    image: qdrant/qdrant:v1.17.0
-#     privileged: true
+    restart: unless-stopped
-#     networks:
+    networks:
-#       - ai-net
+      - ai-net
-#       - pangolin
+      - ai-services
-#     ports:
+    ports:
-#       - "8032:8080"
+      - "6333:6333"
-#     environment:
+      - "6334:6334"
-#       TZ: ${TZ}
+    environment:
-#       HF_TOKEN: ${HF_TOKEN}
+      TZ: ${TZ}
-#       NVIDIA_VISIBLE_DEVICES: all
+      QDRANT__SERVICE__GRPC_PORT: 6334
-#       NVIDIA_DRIVER_CAPABILITIES: compute,utility
+      QDRANT__STORAGE__STORAGE_PATH: /qdrant/storage
-#     volumes:
+      QDRANT__STORAGE__SNAPSHOTS_PATH: /qdrant/snapshots
-#       - /mnt/docker-ssd/docker/appdata/scriberr:/app/data
+    volumes:
-#     deploy:
+      - /mnt/docker-ssd/docker/appdata/qdrant/storage:/qdrant/storage
-#       resources:
+      - /mnt/docker-ssd/docker/appdata/qdrant/snapshots:/qdrant/snapshots
-#         reservations:
+
-#           devices:
+  litellm:
-#             - driver: nvidia
+    container_name: litellm
-#               count: all
+    image: ghcr.io/berriai/litellm:main-latest
-#               capabilities: [gpu]
+    restart: unless-stopped
-#     healthcheck:
+    networks:
-#       test: ["CMD-SHELL", "curl -sf http://127.0.0.1:8080/ || exit 1"]
+      - ai-net
-#       interval: 30s
+      - ai-services
-#       timeout: 10s
+    ports:
-#       retries: 3
+      - "4000:4000"
-#       start_period: 60s
+    environment:
+      TZ: ${TZ}
+      LITELLM_MASTER_KEY: ${LITELLM_MASTER_KEY}
+    volumes:
+      - /mnt/docker-ssd/docker/compose/ai/litellm/config.yaml:/app/config.yaml:ro
+      - /mnt/docker-ssd/docker/appdata/litellm:/app/data
+    command: ["--config", "/app/config.yaml", "--port", "4000"]
+    extra_hosts:
+      - "host.docker.internal:host-gateway"
+
+  reranker:
+    container_name: reranker
+    image: ghcr.io/huggingface/text-embeddings-inference:cpu-latest
+    restart: unless-stopped
+    networks:
+      - ai-net
+      - ai-services
+    ports:
+      - "8787:80"
+    command: ["--model-id", "BAAI/bge-reranker-v2-m3"]
+    volumes:
+      - /mnt/docker-ssd/docker/appdata/reranker:/data
+
+  whisper:
+    container_name: whisper
+    image: fedirz/faster-whisper-server:latest-cuda
+    restart: unless-stopped
+    networks:
+      - ai-net
+      - ai-services
+    ports:
+      - "8786:8000"
+    environment:
+      TZ: ${TZ}
+      WHISPER__MODEL: ${WHISPER_MODEL}
+      WHISPER__DEVICE: cuda
+      WHISPER__COMPUTE_TYPE: float16
+    volumes:
+      - /mnt/docker-ssd/docker/appdata/whisper:/root/.cache/huggingface
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]

ai/litellm/config.yaml

@@ -0,0 +1,99 @@
+model_list:
+  # ---- ROCINANTE (RTX 4090, 24GB) — heavy reasoning ----
+  - model_name: "heavy"
+    litellm_params:
+      model: "ollama/qwen3.6:latest"
+      api_base: "http://10.5.1.112:11434"
+      timeout: 300
+      stream_timeout: 300
+
+  - model_name: "heavy"
+    litellm_params:
+      model: "ollama/qwen2.5:32b"
+      api_base: "http://10.5.1.112:11434"
+      timeout: 300
+      stream_timeout: 300
+
+  - model_name: "heavy"
+    litellm_params:
+      model: "ollama/gemma3:27b"
+      api_base: "http://10.5.1.112:11434"
+      timeout: 300
+      stream_timeout: 300
+
+  # ---- PlausibleDeniability (RTX 2080 Ti, 11GB) — general ----
+  - model_name: "medium"
+    litellm_params:
+      model: "ollama/qwen2.5:14b"
+      api_base: "http://host.docker.internal:11434"
+      timeout: 120
+      stream_timeout: 120
+
+  # ---- N.O.M.A.D. (GTX 1080, 8GB) — lightweight ----
+  - model_name: "light"
+    litellm_params:
+      model: "ollama/deepseek-r1"
+      api_base: "http://10.5.1.16:11434"
+      timeout: 60
+      stream_timeout: 60
+
+  - model_name: "light"
+    litellm_params:
+      model: "ollama/qwen2.5:3b"
+      api_base: "http://10.5.1.16:11434"
+      timeout: 60
+      stream_timeout: 60
+
+  # ---- Embeddings (N.O.M.A.D. has nomic-embed-text) ----
+  - model_name: "embed"
+    litellm_params:
+      model: "ollama/nomic-embed-text:v1.5"
+      api_base: "http://10.5.1.16:11434"
+
+  # ---- Direct model access (bypass tier routing) ----
+  - model_name: "ollama/qwen3.6:latest"
+    litellm_params:
+      model: "ollama/qwen3.6:latest"
+      api_base: "http://10.5.1.112:11434"
+
+  - model_name: "ollama/qwen2.5:32b"
+    litellm_params:
+      model: "ollama/qwen2.5:32b"
+      api_base: "http://10.5.1.112:11434"
+
+  - model_name: "ollama/gemma3:27b"
+    litellm_params:
+      model: "ollama/gemma3:27b"
+      api_base: "http://10.5.1.112:11434"
+
+  - model_name: "ollama/qwen2.5:14b"
+    litellm_params:
+      model: "ollama/qwen2.5:14b"
+      api_base: "http://host.docker.internal:11434"
+
+  - model_name: "ollama/deepseek-r1"
+    litellm_params:
+      model: "ollama/deepseek-r1"
+      api_base: "http://10.5.1.16:11434"
+
+  - model_name: "ollama/qwen2.5:3b"
+    litellm_params:
+      model: "ollama/qwen2.5:3b"
+      api_base: "http://10.5.1.16:11434"
+
+  - model_name: "ollama/nomic-embed-text:v1.5"
+    litellm_params:
+      model: "ollama/nomic-embed-text:v1.5"
+      api_base: "http://10.5.1.16:11434"
+
+litellm_settings:
+  drop_params: true
+  set_verbose: false
+  request_timeout: 300
+  num_retries: 2
+  retry_after: 5
+  allowed_fails: 3
+  cooldown_time: 60
+
+general_settings:
+  master_key: "os.environ/LITELLM_MASTER_KEY"