CONTINUOUS IMPROVEMENT

Release Notes
& Update History

13 Scripts
150 Releases
Jan 26 Last Updated

Recent Updates

Showing 13 of 150 releases

January 2026

Single Image Edit V4, Google Veo 3.1 models, Generate Similar V3, Upscaling + Retouching V2

New
Custom Workflows Folder Access: Added a "Custom Workflows" section in Settings → AI Generation with a button to quickly open the custom workflows folder
Fix
Fixed AI generation progress showing 0% instead of approximate estimation during initial processing

December 2025

Layered output support, new WAN v2.6 models, Flux 2 Max, and SVG paste

New
Layered output support: AI models that produce multiple layers (Qwen Image Layered) now show violet layer badges, support AE export as stacked layers, and include fullscreen layer navigation
Improved
SVG clipboard paste: Paste SVG content directly from clipboard as .svg files with automatic security sanitization
Fix
Fullscreen audio preview now shows waveform and playback controls

New workflows, performance optimizations

New
New WAN Move workflow (WAN-2.1): animate images along motion trajectories captured from After Effects layers
New
New Video Upscale v2 workflow with HD magnification and optional frame interpolation (2x/4x). Frame Interpolation control added to prompt bottom bar
New
New Upscale Image (SeedVR2.5) workflow with single image input
Improved
AI Models modal: workflow cache prevents repeated disk reads when switching tabs
Improved
Workflow loading optimized to only reload on meaningful changes (visibility, paths)
Fix
Fixed "Split to..." failing on videos with timecode data streams (e.g., DaVinci Resolve exports)
Fix
Fixed "Extract frame" failing on videos with non-zero start times (e.g., AI-generated videos)
Fix
Improved RunningHub error messages
Fix
Fixed infinite retry loop in batch pipeline when segments fail (now waits for user action)
Fix
Fixed batch pipeline failing on last segment when audio duration doesn't divide evenly

Item rotation, drag-to-pan preview, Extend Image workflows, and UI improvements

New
New Extend Image (Zoom Out) workflow: expand image borders with AI-generated context using Flux1 Fill Dev
New
Board items can now be rotated via corner handles, with Shift to snap to 15-degree increments
New
Fullscreen media preview now supports drag-to-pan when zoomed in (grab cursor appears, "Drag to pan" hint shown)
Improved
Compare preview navigation now shows image/video thumbnails instead of text labels
Improved
Prompt bottom control dropdowns now close when clicking outside (including on the board)
Improved
Prompt bottom control dropdowns now align to the right edge of buttons for better positioning
Improved
Extend Image (Fast) workflow now has aspect ratio selector with FullHD presets (16:9, 9:16, 1:1)
Fix
Audio input support for fal.ai models (Kling Avatar V2/V2 Pro) - audio_url parameter was not being sent to the API

Clips Pipeline, new Kling models, mouse-anchored zoom, and video merge/split tools

New
New AI Models: Seedream 4.5, Kling 2.6 Pro (T2V/I2V), Kling 2.5 Turbo Pro, Kling Avatar V2/V2 Pro for talking avatar videos
New
Clips Pipeline: batch pipeline for generating video clips from audio segments with automatic video-to-audio sync and concatenation
New
Mouse-position anchored zoom: wheel zoom now keeps the point under cursor fixed (like Figma/Miro)
New
Auto Split (scenes): automatically split videos at scene changes using FFmpeg scene detection
New
Merge Videos: combine multiple selected videos into a single video file
New
Add Sequentially: add selected videos/audio to AE composition one after another (Shift+Click creates new comp)
Fix
FilesModal grid virtualization: improved layout calculation with proper gap handling
Fix
Board deletion: now switches to previous board by creation order
Fix
Auto-save: skip saving while user is dragging or panning
Fix
Audio waveform player: sizing now accounts for board zoom level
Loading more releases…
View Complete Release Archive

Browse all 150 releases across all scripts

Release Notes - Motion Design School Scripts — MDS Scripts