Update README.md

Added instructions for creating .env files and fixed SearXNG setup.

README.md

@@ -20,7 +20,7 @@ Portal at `localai.mydomain.com` switches between stacks to share VRAM.
 ## Directory structure
 
 ```
-~/docker
+homelab-ai/
 ├── ai-image-gen/
 │   ├── docker-compose.yml     ← InvokeAI
 │   └── archive.sh             ← saves Docker images for offline restore
@@ -83,7 +83,12 @@ mkdir -p ~/docker/ai-image-gen/data/invokeai
 mkdir -p ~/docker/ai-llm/data/ollama
 mkdir -p ~/docker/ai-llm/data/open-webui
 mkdir -p ~/docker/ai-llm/data/searxng
-mkdir -p ~/docker/ai-portal
+mkdir -p ~/docker/ai-portal/templates
+```
+
+### 7. Create .env files
+```bash
+echo "TZ=America/New_York" | tee ~/docker/ai-llm/.env ~/docker/ai-image-gen/.env ~/docker/ai-portal/.env
 ```
 
 ### 7. Copy files from repo
@@ -120,7 +125,97 @@ docker exec ai-llm-ollama-1 ollama pull llava:7b            # vision/screenshots
 
 First model for InvokeAI — go to `images.mydomain.com` → Model Manager → search `sd-1.5-inpainting` → Install.
 
-## Open WebUI web search setup
+## InvokeAI Usage Guide
+
+### Installing your first model
+1. Open `https://images.mydomain.com`
+2. Click **Model Manager** (left sidebar)
+3. Click **HuggingFace**
+4. Search: `runwayml/stable-diffusion-inpainting`
+5. Click Install — downloads automatically
+
+### Text to Image (like ChatGPT image generation)
+1. Click the **canvas icon** (left sidebar)
+2. Make sure model is selected at top (e.g. `sd-1.5-inpainting`)
+3. Type your prompt in the prompt box
+4. Click **Invoke**
+5. Results appear in the gallery (right side)
+
+Key settings to learn:
+- **Steps** — higher = more refined, slower (20-30 is good)
+- **CFG Scale** — how closely to follow the prompt (7-9 is good)
+- **Seed** — lock it to reproduce the exact same image
+
+### Inpainting (fixing details in an existing image)
+1. Open an image from the gallery or upload one
+2. Click **Send to Canvas**
+3. Select the **Brush** tool
+4. Paint a mask over the area you want to change (face, hand, background)
+5. Type what you want in that area in the prompt
+6. Adjust **Denoising Strength** — lower (0.3-0.5) = subtle, higher (0.7+) = more creative
+7. Click **Invoke**
+8. The masked area is regenerated, rest of image stays intact
+
+Tips:
+- Use **Inpaint at Full Resolution** for face/hand details
+- Keep denoising low (0.3-0.4) for seamless blending
+- Mask blur of 4-8px softens the edge between inpainted and original
+
+### Using an image as a base (img2img)
+1. Upload or select an image from gallery
+2. Click **Send to Canvas**
+3. Change mode to **Image to Image**
+4. Type your prompt describing what you want
+5. Adjust **Denoising Strength**:
+   - Low (0.3) = keeps original, subtle changes
+   - High (0.8) = major transformation
+6. Click **Invoke**
+
+### Consistent character images (for product lines)
+To maintain a consistent character across many images:
+
+**Method 1 — Seed locking:**
+- Generate until you get the character you want
+- Note the seed number
+- Lock the seed — same seed = same character base every time
+- Change only the prompt details (pose, background, clothing)
+
+**Method 2 — Reference image + low denoising:**
+- Use your established character image as the base
+- Set denoising to 0.3-0.4
+- Change prompt to describe the new scene/pose
+- Character features stay consistent, context changes
+
+**Method 3 — LoRA models:**
+- Train or download a LoRA for your specific character
+- Load the LoRA in Model Manager
+- Add the LoRA trigger word to your prompt
+- Most consistent method for commercial product lines
+- LoRAs available at civitai.com
+
+### Workflow for best results
+1. Generate several variations (change seed each time)
+2. Pick the best base image
+3. Lock the seed
+4. Inpaint problem areas (faces, hands, backgrounds)
+5. Use img2img for scene variations with same character
+6. Upscale final image for print quality
+
+SearXNG's wikidata engine fails on startup due to external blocks. Disable it:
+
+```bash
+# Find the line number
+sudo grep -n "name: wikidata" ~/docker/ai-llm/data/searxng/settings.yml
+
+# Add inactive: true after the engine line (replace 771 with actual line number)
+sudo sed -i '771a\    inactive: true' ~/docker/ai-llm/data/searxng/settings.yml
+
+# Verify
+sudo sed -n '769,774p' ~/docker/ai-llm/data/searxng/settings.yml
+
+# Restart
+docker compose -f ~/docker/ai-llm/docker-compose.yml restart searxng
+```
 In Open WebUI Admin Panel → Settings → Web Search:
 - Engine: Brave
 - API Key: get free key at `api.search.brave.com` (free $5/month credits, set spending cap to Free)

ai-llm/docker-compose.yml

@@ -0,0 +1,74 @@
+# ~/docker/ai-llm/docker-compose.yml
+#
+# ═══════════════════════════════════════════════════════════════════════════════
+# USAGE
+# ═══════════════════════════════════════════════════════════════════════════════
+#
+# Before starting this stack, stop InvokeAI via the portal.
+#
+# Start:
+#   cd ~/docker/ai-llm && docker compose up -d
+#
+# Pull models (run once per model):
+#   docker exec ai-llm-ollama-1 ollama pull qwen2.5-coder:7b
+#   docker exec ai-llm-ollama-1 ollama pull qwen2.5:7b
+#   docker exec ai-llm-ollama-1 ollama pull llava:7b
+#
+# Open WebUI → https://chat.mydomain.com
+#
+# After starting, enable web search in Open WebUI:
+#   Settings → Web Search → Enable → SearXNG → http://searxng:8080
+# ═══════════════════════════════════════════════════════════════════════════════
+
+services:
+
+  # ── Ollama ───────────────────────────────────────────────────────────────────
+  ollama:
+    image: ollama/ollama:latest
+    pull_policy: missing
+    environment:
+      - TZ=America/New_York
+    volumes:
+      - ./data/ollama:/root/.ollama
+    ports:
+      - "11434:11434"
+    deploy:
+      resources:
+        reservations:
+          devices:
+            - driver: nvidia
+              count: 1
+              capabilities: [gpu]
+    restart: unless-stopped
+
+  # ── Open WebUI ───────────────────────────────────────────────────────────────
+  open-webui:
+    image: ghcr.io/open-webui/open-webui:user0
+    pull_policy: missing
+    volumes:
+      - ./data/open-webui:/app/backend/data
+    ports:
+      - "3000:8080"
+    environment:
+      - TZ=America/New_York
+      - OLLAMA_BASE_URL=http://ollama:11434
+      - SEARXNG_QUERY_URL=http://searxng:8080/search?q=<query>&format=json
+    depends_on:
+      - ollama
+      - searxng
+    restart: unless-stopped
+
+  # ── SearXNG ──────────────────────────────────────────────────────────────────
+  # Privacy-respecting meta search — no API key needed
+  # Open WebUI Settings → Web Search → SearXNG → http://searxng:8080
+  searxng:
+    image: searxng/searxng:latest
+    pull_policy: missing
+    volumes:
+      - ./data/searxng:/etc/searxng
+    environment:
+      - TZ=America/New_York
+      - SEARXNG_BASE_URL=http://searxng:8080
+    expose:
+      - 8080
+    restart: unless-stopped

ai-portal/Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.12-slim
+
+# Install docker compose CLI so the app can control other stacks
+RUN apt-get update && apt-get install -y curl && \
+    curl -fsSL https://get.docker.com | sh && \
+    pip install flask && \
+    rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+COPY app.py .
+COPY templates/ templates/
+
+CMD ["python", "app.py"]

ai-portal/app.py

@@ -0,0 +1,117 @@
+import os
+import subprocess
+import threading
+from flask import Flask, render_template, redirect, jsonify
+
+app = Flask(__name__)
+
+IMAGE_STACK = os.environ.get("IMAGE_STACK", "/docker/ai-image-gen")
+LLM_STACK   = os.environ.get("LLM_STACK",   "/docker/ai-llm")
+
+# Track current active stack and swap status
+state = {
+    "active": None,       # "images" | "chat" | None
+    "swapping": False,
+    "message": "No stack active"
+}
+
+def run_compose(stack_dir, command):
+    """Run a docker compose command in a stack directory."""
+    subprocess.run(
+        ["docker", "compose", "-f", f"{stack_dir}/docker-compose.yml"] + command,
+        capture_output=True
+    )
+
+def is_stack_running(stack_dir):
+    """Check if the GPU container for this stack is running."""
+    container = "ai-image-gen-invokeai-1" if "image" in stack_dir else "ai-llm-ollama-1"
+    result = subprocess.run(
+        ["docker", "inspect", "--format", "{{.State.Running}}", container],
+        capture_output=True,
+        text=True
+    )
+    return result.stdout.strip() == "true"
+
+def do_swap(target):
+    """Stop the GPU-heavy container of inactive stack, start target."""
+    state["swapping"] = True
+
+    if target == "images":
+        state["message"] = "Stopping Ollama to free VRAM..."
+        subprocess.run(["docker", "stop", "ai-llm-ollama-1"], capture_output=True)
+        state["message"] = "Starting InvokeAI..."
+        subprocess.run(
+            ["docker", "compose", "-f", f"{IMAGE_STACK}/docker-compose.yml", "up", "-d"],
+            capture_output=True
+        )
+    else:
+        state["message"] = "Stopping InvokeAI to free VRAM..."
+        subprocess.run(["docker", "stop", "ai-image-gen-invokeai-1"], capture_output=True)
+        state["message"] = "Starting Ollama..."
+        subprocess.run(["docker", "start", "ai-llm-ollama-1"], capture_output=True)
+
+    state["active"] = target
+    state["swapping"] = False
+    state["message"] = f"{target.capitalize()} is ready."
+
+@app.route("/")
+@app.route("/portal")
+def portal():
+    image_running = is_stack_running(IMAGE_STACK)
+    llm_running   = is_stack_running(LLM_STACK)
+    return render_template(
+        "portal.html",
+        active=state["active"],
+        swapping=state["swapping"],
+        message=state["message"],
+        image_running=image_running,
+        llm_running=llm_running
+    )
+
+@app.route("/switch/<target>")
+def switch(target):
+    if target not in ("images", "chat"):
+        return "Invalid target", 400
+
+    if state["swapping"]:
+        return redirect("/portal")
+
+    # Warn if same stack requested
+    if state["active"] == target:
+        urls = {"chat": "https://chat.mydomain.com", "images": "https://images.mydomain.com"}
+        return redirect(urls[target])
+
+    # Run swap in background so browser gets immediate response
+    thread = threading.Thread(target=do_swap, args=(target,))
+    thread.daemon = True
+    thread.start()
+
+    return render_template(
+        "switching.html",
+        target=target,
+        message=f"Switching to {target}... this takes about 10 seconds."
+    )
+
+@app.route("/chat")
+def chat():
+    return redirect(URLS["chat"])
+
+@app.route("/images")
+def images():
+    return redirect(URLS["images"])
+
+@app.route("/status")
+def status():
+    """Polled by the switching page to know when swap is done."""
+    return jsonify(state)
+
+if __name__ == "__main__":
+    # Detect what's already running on startup
+    if is_stack_running(IMAGE_STACK):
+        state["active"] = "images"
+        state["message"] = "Images stack is active."
+    elif is_stack_running(LLM_STACK):
+        state["active"] = "chat"
+        state["message"] = "Chat stack is active."
+
+    app.run(host="0.0.0.0", port=8080)

ai-portal/docker-compose.yml

@@ -0,0 +1,26 @@
+# ~/docker/ai-portal/docker-compose.yml
+#
+# Landing page and stack swap controller.
+# This stays running permanently — it is lightweight (no GPU).
+#
+# Start:
+#   cd ~/docker/ai-portal && docker compose up -d
+#
+# Portal → http://192.0.2.1:8080
+# Caddy proxies localai.mydomain.com → here
+
+services:
+
+  portal:
+    build: .
+    ports:
+      - "8080:8080"
+    volumes:
+      # Needs access to Docker socket to start/stop other stacks
+      - /var/run/docker.sock:/var/run/docker.sock
+      # Needs access to the other compose files
+      - /home/user/docker:/docker
+    environment:
+      - IMAGE_STACK=/docker/ai-image-gen
+      - LLM_STACK=/docker/ai-llm
+    restart: unless-stopped

ai-portal/templates/portal.html

@@ -0,0 +1,75 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8">
+  <meta name="viewport" content="width=device-width, initial-scale=1.0">
+  <meta http-equiv="Cache-Control" content="no-cache, no-store, must-revalidate">
+  <meta http-equiv="Pragma" content="no-cache">
+  <meta http-equiv="Expires" content="0">
+  <title>Local AI Portal</title>
+  <style>
+    * { box-sizing: border-box; margin: 0; padding: 0; }
+    body {
+      font-family: system-ui, sans-serif;
+      background: #0f0f0f;
+      color: #e0e0e0;
+      min-height: 100vh;
+      display: flex;
+      align-items: center;
+      justify-content: center;
+    }
+    .card {
+      background: #1a1a1a;
+      border: 1px solid #2a2a2a;
+      border-radius: 16px;
+      padding: 2.5rem;
+      width: 420px;
+      text-align: center;
+    }
+    h1 { font-size: 1.4rem; font-weight: 500; margin-bottom: 0.4rem; }
+    .status { font-size: 0.85rem; color: #888; margin-bottom: 2rem; }
+    .buttons { display: flex; gap: 1rem; }
+    .btn {
+      flex: 1;
+      padding: 1rem;
+      border-radius: 12px;
+      border: 1px solid #333;
+      background: #222;
+      color: #e0e0e0;
+      font-size: 1rem;
+      cursor: pointer;
+      text-decoration: none;
+      display: block;
+      transition: background 0.15s;
+    }
+    .btn:hover { background: #2a2a2a; }
+    .btn.active { border-color: #4a9eff; background: #1a2a3a; color: #4a9eff; }
+    .btn.active-green { border-color: #4aff8a; background: #1a3a2a; color: #4aff8a; }
+    .btn span { display: block; font-size: 0.75rem; color: #666; margin-top: 0.3rem; }
+    .btn.active span, .btn.active-green span { color: inherit; opacity: 0.7; }
+    .message { margin-top: 1.5rem; font-size: 0.8rem; color: #555; }
+    {% if swapping %}
+    .btn { pointer-events: none; opacity: 0.5; }
+    {% endif %}
+  </style>
+</head>
+<body>
+  <div class="card">
+    <h1>Local AI</h1>
+    <p class="status">{{ message }}</p>
+    <div class="buttons">
+      <a href="/switch/chat"
+         class="btn {% if active == 'chat' %}active{% endif %}">
+        💬 Chat
+        <span>{% if active == 'chat' %}active{% else %}Open WebUI + Ollama{% endif %}</span>
+      </a>
+      <a href="/switch/images"
+         class="btn {% if active == 'images' %}active-green{% endif %}">
+        🎨 Images
+        <span>{% if active == 'images' %}active{% else %}InvokeAI{% endif %}</span>
+      </a>
+    </div>
+    <p class="message">{{ message }}</p>
+  </div>
+</body>
+</html>

ai-portal/templates/switching.html

@@ -0,0 +1,67 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+  <meta charset="UTF-8">
+  <title>Switching...</title>
+  <style>
+    * { box-sizing: border-box; margin: 0; padding: 0; }
+    body {
+      font-family: system-ui, sans-serif;
+      background: #0f0f0f;
+      color: #e0e0e0;
+      min-height: 100vh;
+      display: flex;
+      align-items: center;
+      justify-content: center;
+    }
+    .card {
+      background: #1a1a1a;
+      border: 1px solid #2a2a2a;
+      border-radius: 16px;
+      padding: 2.5rem;
+      width: 420px;
+      text-align: center;
+    }
+    h1 { font-size: 1.2rem; font-weight: 500; margin-bottom: 1rem; }
+    .message { color: #888; font-size: 0.9rem; margin-bottom: 1.5rem; }
+    .spinner {
+      width: 32px; height: 32px;
+      border: 3px solid #333;
+      border-top-color: #4a9eff;
+      border-radius: 50%;
+      animation: spin 0.8s linear infinite;
+      margin: 0 auto;
+    }
+    @keyframes spin { to { transform: rotate(360deg); } }
+    .status { margin-top: 1rem; font-size: 0.8rem; color: #555; }
+  </style>
+</head>
+<body>
+  <div class="card">
+    <h1>Switching to {{ target }}...</h1>
+    <p class="message">{{ message }}</p>
+    <div class="spinner"></div>
+    <p class="status" id="status">Please wait...</p>
+  </div>
+  <script>
+    const target = "{{ target }}";
+    function poll() {
+      fetch("/status")
+        .then(r => r.json())
+        .then(data => {
+          document.getElementById("status").textContent = data.message;
+          if (!data.swapping && data.active === target) {
+            const urls = { chat: "https://chat.mydomain.com", images: "https://images.mydomain.com" };
+            // Extra delay for images stack — InvokeAI takes longer to be ready
+            const delay = target === "images" ? 8000 : 2000;
+            setTimeout(() => { window.location.href = urls[target]; }, delay);
+          } else {
+            setTimeout(poll, 1500);
+          }
+        })
+        .catch(() => setTimeout(poll, 2000));
+    }
+    poll();
+  </script>
+</body>
+</html>