Update Docker

Squash commit that merges #43, #44, and #45

Create .dockerignore

Make compose marginally better

Un-scuffed the Dockerfile

.dockerignore

@@ -0,0 +1 @@
+models/

Dockerfile

@@ -10,8 +10,8 @@ RUN apt-get update && apt-get install -y \
 # Set the working directory in the container
 WORKDIR /usr/src/app
 
-# Copy the current directory contents into the container at /usr/src/app
-COPY . .
+# Get requirements
+COPY requirements.txt requirements.txt
 
 # Install torch with CUDA support and exllamav2
 RUN pip install torch --extra-index-url https://download.pytorch.org/whl/cu121
@@ -20,14 +20,11 @@ RUN pip install exllamav2
 # Install any other needed packages specified in requirements.txt
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy the sample config file to the main config
-RUN cp config_sample.yml config.yml
+# Copy the current directory contents into the container at /usr/src/app
+COPY . .
 
 # Make port 5000 available to the world outside this container
 EXPOSE 5000
 
-# Define environment variable
-ENV NAME World
-
 # Run main.py when the container launches
 CMD ["python3", "main.py"]

docker-compose.yml

@@ -7,9 +7,15 @@ services:
         INSTALL_FSCHAT: "true"  # Set this to "true" or "false" as needed
     ports:
       - "5000:5000"
-    runtime: nvidia
     environment:
       - NAME=TabbyAPI
       - NVIDIA_VISIBLE_DEVICES=all
     volumes:
-      - /mnt/nvme/models:/usr/src/app/models
+      - ./models:/usr/src/app/models
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: all
+              capabilities: [gpu]