Generate Media Button
The Generate Media button toggles between chat mode and media generation mode, giving you access to AI-powered video and image generation tools.
Where to Find It
The Generate Media button is located in the Composer toolbar, to the right of the web search button.
Chat Mode (Disabled)
When Generate Media is off:
Text input for conversations
File uploads work normally
AI assistant responds to questions
When Generate Media is on:
Model Selector appears - Choose video or image models
Media Settings Pills appear - Aspect ratio, resolution, duration, audio
Generation controls become active
Interface optimized for media creation
Click the Generate Media toggle/button in the composer
Media controls appear above the text input
Select your model from the model selector
Adjust settings (aspect ratio, duration, etc.)
Enter your prompt and generate
Click the toggle again to return to chat mode
Media controls hide automatically
Standard chat interface returns
When Generate Media is enabled, you'll see:
Video Models: Sora 2, Veo 3.1, Kling, Hailuo, WAN, etc.
Image Models: Flux 2 Max, GPT Image 1.5, Nano Banana Pro, etc.
Video-to-Video: SAM 3, Kling VFX, LTX Reshoot (when video attached)
Aspect Ratio Pills
16:9 - Landscape (YouTube, general video)
9:16 - Vertical (TikTok, Instagram Reels, YouTube Shorts)
1:1 - Square (Instagram posts)
4:3 - Classic/vintage look
Click the chevron up to expand the menu to see more options (options vary per model)
Resolution Settings
720p - Faster generation, lower cost
1080p - Standard quality (most models)
4K - Highest quality (Sora 2 Pro, WAN 2.5)
Duration Settings (Video Only)
4 seconds - Quick clips, social media
8 seconds - Standard length (most models)
10 seconds - Longer clips (Kling, Sora)
12 seconds - Maximum (Sora 2 only)
Audio Toggle (Video Only)
Enable - For models that support audio (Veo 3.1, Kling 3.0)
Disable - Silent generation (add audio in post)
Automatic Mode Detection
The Generate Media interface automatically adapts:
No images attached → Text-to-video models available
Enter text prompt → Generate video
1 image attached → Image-to-video models available
Enter motion description → Animate the image
Transition Mode
2 images attached → Transition models available
Describe camera movement → Create smooth transition
1 video attached → Video editing models available (SAM 3, Kling VFX, etc.)
Enter edit description → Modify the video
Workflow Examples
Example 1: Generate a Video
Select video model (e.g., Veo 3.1)
Set aspect ratio (e.g., 16:9)
Set duration (e.g., 8 seconds)
Enable audio (if desired)
Enter prompt: "A cinematic shot of a mountain range at sunset."
Click Send or press Enter
Wait for the generation to complete
Example 2: Create an Image
Switch to image mode (if needed)
Select image model (e.g., Flux 2 Max)
Set aspect ratio (e.g., 1:1)
Enter prompt: "A modern coffee shop interior, warm lighting."
Image appears in the chat
Example 3: Animate an Image
Image-to-video models automatically appear
Select model (e.g., Veo 3.1)
Enter motion description: "Slow push-in with camera movement."
Tips for Best Results
Start simple - Basic prompts work well, then iterate
Be specific - More details = better results
Match model to use case - See model comparison guides
Adjust settings - Different aspect ratios and durations affect output
Iterate - Generate, review, refine, regenerate
Troubleshooting
Check you're in the Composer area (bottom of panel)
Verify the panel is fully loaded
Try refreshing Premiere Pro
Ensure Generate Media is enabled (toggle should be on)
Check you have a valid Fal.ai API key
Verify you have credits in your Fal.ai account
"Model selector is empty."
Checkthat your Fal.ai account has sufficient credits
Verify the API key is correct. The y configured
Some models may have temporary availability issues
"Generation fails"
Check your Fal.ai balance
Verify internet connection
Try a shorter prompt or different model
Check model-specific requirements
Next: Learn about the Frame Capture Button to export frames from your Premiere Pro timeline.