Runway Gen-3 Alpha is one of the most powerful AI video generation tools available. This tutorial walks you through everything you need to know to start creating professional-quality AI videos.
Getting Started with Runway
Step 1: Create an Account
- Visit runwayml.com
- Click “Sign Up” and create an account
- Verify your email address
- You’ll receive free credits to start experimenting
Step 2: Navigate to Gen-3 Alpha
- From the dashboard, click “Create”
- Select “Gen-3 Alpha” from the model options
- You’ll see the main generation interface
Text-to-Video Basics
The simplest way to create video is with a text prompt.
Writing Effective Prompts
Gen-3 Alpha responds best to descriptive, cinematic prompts. Follow this structure:
[Camera movement], [Subject description], [Action], [Setting], [Lighting/Mood]
Example prompts:
Cinematic shot:
Slow dolly shot of a woman walking through a misty forest at dawn,
golden light filtering through trees, cinematic, 35mm film
Product shot:
Smooth orbit around a luxury watch on black velvet,
studio lighting with subtle reflections, commercial photography style
Action sequence:
Tracking shot following a motorcycle speeding through neon-lit
Tokyo streets at night, rain falling, cyberpunk atmosphere
Generation Settings
Before generating, configure these settings:
- Duration: 5 or 10 seconds (10 costs more credits)
- Aspect Ratio: 16:9 (landscape), 9:16 (portrait), 1:1 (square)
- Resolution: 720p or 1080p
Click “Generate” and wait 60-120 seconds for your video.
Image-to-Video
Transform still images into dynamic video clips.
How to Use Image-to-Video
- Click the image upload icon
- Select your source image (recommended: 1920x1080 or higher)
- Write a prompt describing the desired motion
- Generate
Best Practices for Image-to-Video
Do:
- Use high-resolution, sharp source images
- Describe the motion, not the image content
- Keep camera movements simple
Don’t:
- Use blurry or low-quality images
- Ask for drastic scene changes
- Expect the subject to completely transform
Example prompt for image-to-video:
Gentle wind blowing through hair, subtle smile forming,
soft camera push in, natural lighting
Camera Controls
Gen-3 Alpha offers powerful camera control options.
Available Camera Movements
| Movement | Description | When to Use |
|---|---|---|
| Static | No camera movement | Dialogue, portraits |
| Pan | Horizontal rotation | Revealing scenes |
| Tilt | Vertical rotation | Tall subjects |
| Dolly | Move toward/away | Dramatic emphasis |
| Truck | Move left/right | Following action |
| Zoom | In/out without moving | Focus attention |
| Orbit | Circle around subject | Product shots |
Using Camera Presets
- Click “Camera Motion” in the interface
- Select a preset movement
- Adjust intensity (subtle, medium, dramatic)
- Combine with your text prompt
Custom Camera Instructions
Include camera directions in your prompt:
Slow push in on a coffee cup with steam rising,
starting from medium shot to close-up,
shallow depth of field
Advanced Techniques
Prompt Weighting
Emphasize important elements by placing them first or repeating key terms:
Close-up portrait, close-up of eyes,
elderly man with weathered face,
deep wrinkles, wise expression,
soft window light
Style References
Reference specific visual styles:
Wes Anderson style, symmetrical composition,
pastel colors, a hotel lobby with
vintage furniture, tracking shot
Negative Prompting
While Gen-3 doesn’t have formal negative prompts, you can guide away from unwanted elements:
Realistic footage, not animated, not CGI,
natural skin texture, no morphing artifacts
Common Issues and Solutions
Problem: Faces look distorted
Solution:
- Keep faces larger in frame
- Use “realistic face, natural features” in prompt
- Try image-to-video with a clear face photo
Problem: Motion is too fast/slow
Solution:
- Specify pacing: “slow motion,” “real-time,” “time-lapse”
- Use camera movement words like “gentle,” “rapid,” “steady”
Problem: Scene changes unexpectedly
Solution:
- Be more specific about the entire scene
- Use image-to-video for more control
- Add “consistent scene, no cuts” to prompt
Problem: Low quality output
Solution:
- Select 1080p output
- Add “high quality, detailed, sharp focus” to prompt
- Use 10-second duration for better coherence
Credit Costs
Understanding credit usage helps budget your projects:
| Setting | Credits per Generation |
|---|---|
| 5 sec, 720p | 50 credits |
| 5 sec, 1080p | 100 credits |
| 10 sec, 720p | 100 credits |
| 10 sec, 1080p | 200 credits |
Tip: Generate at 720p first to test prompts, then create final versions at 1080p.
Workflow Tips
1. Start with Testing
Generate multiple 5-second, 720p versions to refine your prompt before committing to high-quality renders.
2. Build a Prompt Library
Save successful prompts for reuse. Create templates for common shot types:
- Talking head template
- Product showcase template
- Landscape/establishing shot template
- Action sequence template
3. Combine with Editing Software
Gen-3 videos work great as elements in larger projects:
- Generate multiple clips
- Import into Premiere Pro, DaVinci Resolve, or Final Cut
- Color grade, add transitions, and combine
- Add music and sound design
4. Upscale When Needed
Use Runway’s upscaling feature or external tools like Topaz Video AI for higher resolution final output.
Example Project: Product Commercial
Let’s create a 30-second product video:
Shot 1 - Hero reveal (10 sec):
Slow orbit around a sleek smartphone on white surface,
dramatic studio lighting, product photography,
subtle reflections, premium feel
Shot 2 - Detail shot (5 sec):
Extreme close-up of phone camera lens array,
rack focus, metallic texture visible,
macro photography style
Shot 3 - Lifestyle (10 sec):
Tracking shot of young professional using phone
while walking through modern office,
natural lighting, shallow depth of field
Shot 4 - Logo/CTA (5 sec):
Static shot, phone centered on gradient background,
minimal, clean, commercial end card style
Combine these in your editor, add music, and you have a professional commercial.
Next Steps
- Explore AI Video Prompting Guide for advanced prompt techniques
- Compare with Kling AI Tutorial
- Learn about Video-to-Video editing