EricRollei-Local_LLM_Prompt_Enhancer/docs/QUICKSTART.md

Quick Start Guide

Get up and running with AI Video Prompt Expander in 5 minutes!

Step 1: Install LLM Backend

Option A: LM Studio (Recommended for Beginners)

1. Download from https://lmstudio.ai/
2. Install and open LM Studio
3. Go to "Models" tab → Search for "llama-3-8b"
4. Download a model (recommended: Meta-Llama-3-8B-Instruct)
5. Go to "Local Server" tab → Select your model → Click "Start Server"
6. Note the endpoint (default: http://localhost:1234)

Option B: Ollama (Better for Advanced Users)

1. Download from https://ollama.ai/
2. Install Ollama
3. Open terminal/command prompt
4. Run: ollama pull llama3
5. Server starts automatically (endpoint: http://localhost:11434)

Option C: Local Qwen3-VL Vision Backend (Optional)

Use this if you want the image-to-video or image-to-image nodes to caption images without calling LM Studio/Ollama.

1. Install Python packages: pip install transformers accelerate huggingface_hub bitsandbytes
2. Download a Qwen3-VL model (e.g., Qwen/Qwen3-VL-4B-Instruct)
3. In the node, set vision_backend to qwen3_vl
4. Set vision_model_name to the repo or local path (append @4bit/@8bit if desired)
5. Use the vision_endpoint field for optional overrides like quant=4bit;attn=sdpa

Step 2: Verify Installation

1. Open terminal in the video_prompter directory
2. Run: python test_setup.py
3. Choose your LLM backend and test connection
4. You should see "✅ ALL TESTS COMPLETED"

Step 3: Restart ComfyUI

1. If ComfyUI is running, close it
2. Start ComfyUI
3. You should see the node load without errors

Step 4: Add Node to Workflow

1. Right-click in ComfyUI → Add Node
2. Navigate to: video → prompting → AI Video Prompt Expander
3. The node will appear with all inputs ready

Step 5: Basic Usage

Your First Prompt Expansion

Simple Setup:

basic_prompt: "A cat playing piano in a cozy room"
preset: cinematic
expansion_tier: auto
mode: text-to-video
llm_backend: lm_studio (or ollama)
model_name: llama3
api_endpoint: http://localhost:1234/v1 (LM Studio) or http://localhost:11434/api (Ollama)
temperature: 0.7
positive_keywords: (leave empty for now)
negative_keywords: (leave empty for now)
num_variations: 1
save_to_file: true
filename_base: my_first_prompt

Click "Queue Prompt"

The node will:

1. Connect to your LLM
2. Analyze your input
3. Expand it to a detailed cinematic prompt
4. Generate a negative prompt
5. Save everything to a file
6. Show status with file location

Check Your Output

Look at the outputs:

• positive_prompt_1: Your enhanced prompt (connect to video generation)
• negative_prompt: Auto-generated negatives (connect to video generation)
• status: Shows success and file save location

Open the saved file in ComfyUI/output/video_prompts/ to see the full breakdown!

Step 6: Connect to Video Generation

[AI Video Prompt Expander]
    ↓ positive_prompt_1 → [Your Video Gen Node] → prompt
    ↓ negative_prompt → [Your Video Gen Node] → negative_prompt

Now generate your video!

Common First-Time Issues

"Cannot connect to LM Studio"

• Make sure LM Studio's server is running (green indicator)
• Check the endpoint matches (http://localhost:1234/v1)
• Try clicking "Test Connection" in LM Studio

"Cannot connect to Ollama"

• Run ollama serve in terminal if needed
• Check if Ollama is running: ollama list
• Verify endpoint: http://localhost:11434/api

"Model not found"

• In LM Studio: Make sure model is selected in Local Server tab
• In Ollama: Run ollama list to see available models
• Match the model_name exactly as shown

Node doesn't appear in ComfyUI

• Restart ComfyUI completely
• Check for error messages in ComfyUI console
• Verify all files are in custom_nodes/video_prompter/

Next Steps

Once you have basic expansion working:

1. Try Different Presets

   • Change preset to "noir", "action", "surreal"
   • See how dramatically the style changes

2. Experiment with Tiers

   • Compare "basic", "enhanced", "advanced", "cinematic"
   • See the detail increase with each tier

3. Add Keywords

   • If using LoRAs, add triggers to positive_keywords
   • Try style keywords like "professional, high-quality"

4. Generate Variations

   • Set num_variations to 2 or 3
   • Connect each output to different video gen nodes
   • Pick the best result

5. Adjust Temperature

   • Lower (0.3-0.5) for consistent style
   • Higher (0.8-1.2) for creative variations

Example Workflows

Workflow 1: Single Video

[AI Video Prompt Expander]
    ↓ (positive_prompt_1)
[CogVideoX Text2Video]
    ↓
[Save Video]

Workflow 2: Multiple Variations

[AI Video Prompt Expander] (num_variations: 3)
    ↓ (positive_prompt_1) → [Video Gen] → [Save as version_1]
    ↓ (positive_prompt_2) → [Video Gen] → [Save as version_2]
    ↓ (positive_prompt_3) → [Video Gen] → [Save as version_3]

Workflow 3: Image-to-Video

[Load Image]
    ↓
[AI Video Prompt Expander] (mode: image-to-video)
    ↓ (positive_prompt_1)
[Image2Video Node]
    ↓
[Save Video]

Tips for Success

1. Start Simple: Let the node do the work
2. Use Auto Tier: It's smart about detecting complexity
3. Save Prompts: Build a library for reference
4. Try Presets: Each one is dramatically different
5. Read Status: It tells you exactly what happened
6. Check Files: Saved prompts include full breakdown

Resources

• Full Documentation: See README.md
• Configuration Guide: See CONFIGURATION.md
• Presets Reference: See presets.py
• Test Your Setup: Run test_setup.py

Getting Help

If something isn't working:

1. Run python test_setup.py - it will identify issues
2. Check the status output - it shows errors clearly
3. Review your LLM backend connection
4. Check ComfyUI console for error messages
5. Verify model is loaded and endpoint is correct

Success!

If you:

• ✅ See the node in ComfyUI
• ✅ Get a status message like "✅ Generated 1 variation(s)..."
• ✅ Have text in positive_prompt_1 output
• ✅ See a saved file in output/video_prompts/

You're ready to create amazing video prompts! 🎬

Now go make something awesome! 🚀