Skip to content

Conversation

@blendi-remade
Copy link
Owner

Add Image-to-Video Generation with Performance Optimizations
New Features
AI Image-to-Video: Integrated fal.ai's Veo 3.1 fast endpoint for converting timeline images to video animations
Unified AI Panel: Added segmented control to switch between Image and Video generation modes
Timeline Integration: Added "Animate with AI" context menu option for image elements
Preview Quality Control: Implemented dynamic quality settings (25%, 50%, 75%, 100%) to reduce memory usage and improve playback performance
Improvements
Project-Specific AI State: AI generation history and results now properly reset when switching/creating projects
Memory Management: Removed generation history storage and implemented aggressive blob URL cleanup to prevent memory leaks
Video Audio Playback: Fixed audio not playing for videos on the timeline by rendering hidden video elements
Toast Notifications: Aligned AI generation toasts with OpenCut's design system
Technical Changes
Added new types for video generation parameters (duration, resolution, aspect ratio)
Created /api/ai/generate-video endpoint
Enhanced video cache pool size for better performance
Added preview quality state management
Improved frame cache invalidation logic
Build Fixes
Added missing @upstash/redis dependency to auth package
Added missing @radix-ui/react-dialog dependency
Removed output: "standalone" to fix Windows symlink permission issues

@blendi-remade blendi-remade merged commit 84c6cad into main Oct 21, 2025
0 of 3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants