added new docker stack

.env

@@ -1,12 +1,12 @@
 ## Set number of threads.
 ## Note: prefer the number of physical cores. Overbooking the CPU degrades performance notably.
-# THREADS=14
+#THREADS=6
 
 ## Specify a different bind address (defaults to ":8080")
-# ADDRESS=127.0.0.1:8080
+#ADDRESS=192.168.2.101:80
 
 ## Default models context size
-# CONTEXT_SIZE=512
+#CONTEXT_SIZE=512
 #
 ## Define galleries.
 ## models will to install will be visible in `/models/available`
@@ -21,13 +21,13 @@
 MODELS_PATH=/models
 
 ## Enable debug mode
-# DEBUG=true
+DEBUG=true
 
 ## Specify a build type. Available: cublas, openblas, clblas.
-# BUILD_TYPE=openblas
+BUILD_TYPE=cublas
 
 ## Uncomment and set to true to enable rebuilding from source
-# REBUILD=true
+REBUILD=false
 
 ## Enable go tags, available: stablediffusion, tts
 ## stablediffusion: image generation with stablediffusion
@@ -37,7 +37,7 @@ MODELS_PATH=/models
 # GO_TAGS=stablediffusion
 
 ## Path where to store generated images
-# IMAGE_PATH=/tmp
+IMAGE_PATH=/tmp
 
 ## Specify a default upload limit in MB (whisper)
-# UPLOAD_LIMIT
+# UPLOAD_LIMIT

.gitignore

@@ -29,4 +29,14 @@ release/
 # Generated during build
 backend-assets/
 
-/ggml-metal.metal
+/ggml-metal.metal
+.env
+pictures/cc1EUUHn.s
+pictures/cc86q29b.s
+pictures/cch1MqnM.s
+pictures/ccj0ZrSv.s
+pictures/ccl2gcGz.s
+pictures/ccOMMq5C.s
+cuda_12.0.0_525.60.13_linux.run
+myapp
+helloworld.cu

docker-compose.yaml

@@ -1,15 +0,0 @@
-version: '3.6'
-
-services:
-  api:
-    image: quay.io/go-skynet/local-ai:latest
-    build:
-      context: .
-      dockerfile: Dockerfile
-    ports:
-      - 8080:8080
-    env_file:
-      - .env
-    volumes:
-      - ./models:/models:cached
-    command: ["/usr/bin/local-ai" ]

docker-compose.yml

@@ -0,0 +1,40 @@
+version: '3.6'
+
+services:
+  api:
+    container_name: local-ai
+    image: quay.io/go-skynet/local-ai:v0.19.0-cublas-cuda12
+    restart: always
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]
+    ports:
+      - 8080:8080
+    environment:
+        - DEBUG=true
+        - MODELS_PATH=/models
+        - 'PRELOAD_MODELS=[{"url": "github:go-skynet/model-gallery/openllama_7b.yaml", "name": "gpt-3.5-turbo", "overrides": { "f16":true, "gpu_layers": 35, "mmap": true, "batch": 512 } } ]'        
+        - THREADS=6
+        - BUILD_TYPE=cublas
+        - REBUILD=true
+        - NVIDIA_VISIBLE_DEVICES=all
+        - NVIDIA_DRIVER_CAPABILITIES=all
+    volumes:
+        - ./models:/models:cached
+        - ./pictures/:/tmp/
+    command: ["/usr/bin/local-ai"]
+
+  flowise:
+    container_name: flowise-ai
+    image: flowiseai/flowise:latest
+    restart: always
+    ports:
+        - 3000:3000
+    volumes:
+        - ~/.flowise:/root/.flowise
+        - ./logs:/logs/
+    command: /bin/sh -c "sleep 3; flowise start"