ci: Docker workflow to build and slim images, update README for usage instructions

.github/workflows/docker-publish.yml

@@ -1,4 +1,4 @@
-name: Docker Build and Publish
+name: Docker Build, Slim, and Publish
 
 on:
   push:
@@ -35,42 +35,64 @@ jobs:
           echo "GPU_IMAGE_NAME=${{ env.REGISTRY }}/$(echo ${{ env.IMAGE_NAME }} | tr '[:upper:]' '[:lower:]')-gpu" >> $GITHUB_ENV
           echo "CPU_IMAGE_NAME=${{ env.REGISTRY }}/$(echo ${{ env.IMAGE_NAME }} | tr '[:upper:]' '[:lower:]')-cpu" >> $GITHUB_ENV
           echo "UI_IMAGE_NAME=${{ env.REGISTRY }}/$(echo ${{ env.IMAGE_NAME }} | tr '[:upper:]' '[:lower:]')-ui" >> $GITHUB_ENV
-          echo "VERSION=${GITHUB_REF#refs/tags/}" >> $GITHUB_ENV
 
-      # Extract version tag if it exists
-      - name: Set version tag
-        if: startsWith(github.ref, 'refs/tags/')
-        run: echo "VERSION=${GITHUB_REF#refs/tags/}" >> $GITHUB_ENV
-
-      # Build and push GPU version
-      - name: Build and push GPU Docker image
+      # Build GPU version
+      - name: Build GPU Docker image
         uses: docker/build-push-action@v5
         with:
           context: .
           file: ./docker/gpu/Dockerfile
-          push: true
-          tags: |
-            ${{ env.GPU_IMAGE_NAME }}:latest
-            ${{ env.GPU_IMAGE_NAME }}:${{ env.VERSION }}
+          push: false
+          load: true
+          tags: ${{ env.GPU_IMAGE_NAME }}:v0.1.0
           build-args: |
             DOCKER_BUILDKIT=1
           platforms: linux/amd64
 
-      # Build and push CPU version
-      - name: Build and push CPU Docker image
+      # Slim GPU version
+      - name: Slim GPU Docker image
+        uses: kitabisa/docker-slim-action@v1
+        env:
+          DSLIM_HTTP_PROBE: false
+        with:
+          target: ${{ env.GPU_IMAGE_NAME }}:v0.1.0
+          tag: v0.1.0-slim
+
+      # Push GPU versions
+      - name: Push GPU Docker images
+        run: |
+          docker push ${{ env.GPU_IMAGE_NAME }}:v0.1.0
+          docker push ${{ env.GPU_IMAGE_NAME }}:v0.1.0-slim
+
+      # Build CPU version
+      - name: Build CPU Docker image
         uses: docker/build-push-action@v5
         with:
           context: .
           file: ./docker/cpu/Dockerfile
-          push: true
-          tags: |
-            ${{ env.CPU_IMAGE_NAME }}:latest
-            ${{ env.CPU_IMAGE_NAME }}:${{ env.VERSION }}
+          push: false
+          load: true
+          tags: ${{ env.CPU_IMAGE_NAME }}:v0.1.0
           build-args: |
             DOCKER_BUILDKIT=1
           platforms: linux/amd64
 
-      # Build and push UI version
+      # Slim CPU version
+      - name: Slim CPU Docker image
+        uses: kitabisa/docker-slim-action@v1
+        env:
+          DSLIM_HTTP_PROBE: false
+        with:
+          target: ${{ env.CPU_IMAGE_NAME }}:v0.1.0
+          tag: v0.1.0-slim
+
+      # Push CPU versions
+      - name: Push CPU Docker images
+        run: |
+          docker push ${{ env.CPU_IMAGE_NAME }}:v0.1.0
+          docker push ${{ env.CPU_IMAGE_NAME }}:v0.1.0-slim
+
+      # Build and push UI version (unchanged)
       - name: Build and push UI Docker image
         uses: docker/build-push-action@v5
         with:
@@ -78,8 +100,7 @@ jobs:
           file: ./ui/Dockerfile
           push: true
           tags: |
-            ${{ env.UI_IMAGE_NAME }}:latest
-            ${{ env.UI_IMAGE_NAME }}:${{ env.VERSION }}
+            ${{ env.UI_IMAGE_NAME }}:v0.1.0
           build-args: |
             DOCKER_BUILDKIT=1
           platforms: linux/amd64

README.md

@@ -24,7 +24,7 @@ Dockerized FastAPI wrapper for [Kokoro-82M](https://huggingface.co/hexgrad/Kokor
 The service can be accessed through either the API endpoints or the Gradio web interface.
 
 1. Install prerequisites:
-   - Install [Docker Desktop](https://www.docker.com/products/docker-desktop/) + [Git](https://git-scm.com/downloads)
+   - Install [Docker Desktop](https://www.docker.com/products/docker-desktop/)
    - Clone the repository:
         ```bash
         git clone https://github.com/remsky/Kokoro-FastAPI.git
@@ -33,20 +33,20 @@ The service can be accessed through either the API endpoints or the Gradio web i
 
 2. Start the service:
    
-   a. Using Docker Compose (recommended for full setup including UI):
+   - Using Docker Compose (Full setup including UI):
         ```bash
         docker compose up --build # for GPU
-        #docker compose -f docker-compose.cpu.yml up --build # for CPU
+        docker compose -f docker-compose.cpu.yml up --build # for CPU
         ```
-   
-   b. Running the API alone using Docker:
+   - OR running the API alone using Docker (model + voice packs baked in):
         ```bash
-        # For CPU version
-        docker run -p 8880:8880 kokoro-fastapi-cpu
-        
-        # For GPU version (requires NVIDIA Container Toolkit)
-        docker run --gpus all -p 8880:8880 kokoro-fastapi-gpu
+
+        docker run -p 8880:8880 ghcr.io/remsky/kokoro-fastapi-cpu:latest # CPU
+        docker run --gpus all -p 8880:8880 ghcr.io/remsky/kokoro-fastapi-gpu:latest # Nvidia GPU
+        # Minified versions are available with `:latest-slim` tag.
         ```
+        
+        
 2. Run locally as an OpenAI-Compatible Speech Endpoint
     ```python
     from openai import OpenAI