chore(deploy): one-node mode uses store image with embedded qdrant

Author: Thomas Tupper

README.md

@@ -3,7 +3,7 @@
 Deployment pack for FoxMemory (Node/TS services + Mem0 OSS-compatible store).
 
 ## Modes
-1. `compose.one.yml` — one-node (qdrant + infer + store)
+1. `compose.one.yml` — one-node (infer + store with embedded qdrant)
 2. `compose.split.yml` — split topology for Mini/R720 style deployments
 
 ## Quick start
@@ -32,4 +32,4 @@ curl -s -X POST http://localhost:8082/v1/memories/search \
 ## Notes
 - Uses Docker Hub images by default.
 - Configure OPENAI_API_KEY if using OpenAI LLM/embedder path.
-- Qdrant persists in named Docker volume.
+- In one-node mode, Qdrant is embedded in store and persisted via `foxmemory_store_data`.

compose.one.yml

@@ -1,12 +1,4 @@
 services:
-  qdrant:
-    image: qdrant/qdrant:v1.13.2
-    container_name: foxmemory-qdrant
-    ports:
-      - "6333:6333"
-    volumes:
-      - qdrant_data:/qdrant/storage
-
   infer:
     image: ${FOXMEMORY_INFER_IMAGE}
     container_name: foxmemory-infer
@@ -21,17 +13,20 @@ services:
     container_name: foxmemory-store
     ports:
       - "8082:8082"
+      - "6333:6333"
     environment:
       - OPENAI_API_KEY=${OPENAI_API_KEY}
       - MEM0_LLM_MODEL=${MEM0_LLM_MODEL}
       - MEM0_EMBED_MODEL=${MEM0_EMBED_MODEL}
-      - QDRANT_HOST=qdrant
+      - QDRANT_HOST=127.0.0.1
       - QDRANT_PORT=6333
       - QDRANT_COLLECTION=${QDRANT_COLLECTION}
       - MEM0_HISTORY_DB_PATH=/tmp/history.db
+      - QDRANT_STORAGE_PATH=/qdrant/storage
+    volumes:
+      - foxmemory_store_data:/qdrant/storage
     depends_on:
-      - qdrant
       - infer
 
 volumes:
-  qdrant_data:
+  foxmemory_store_data: