v1.1.5

10x Faster: Major Performance Improvements for Bulk Editing

We’ve just shipped a major performance update that makes bulk editing up to 10x faster for large design systems! 🚀

The Problem

Many of you manage design systems with hundreds or even thousands of tokens. We heard your feedback: editing large token sets was getting sluggish, especially when:

  • Updating multiple variables at once
  • Working with color shades (50+ tokens per palette)
  • Bulk importing from JSON files

What We Fixed

1. Optimized Variable Updates (5x faster)

We rewrote our variable update logic to batch Figma API calls. Instead of making individual updates, we now:

  • Group updates into batches of 100
  • Use async processing
  • Cache unchanged values

Result: Updating 500 tokens now takes ~2 seconds instead of ~10 seconds.

2. Smarter Re-rendering (3x faster UI)

The spreadsheet view now uses virtual scrolling and smart diffing:

  • Only visible rows are rendered
  • Changed cells update individually
  • Scroll performance is buttery smooth even with 1000+ rows

3. JSON Import Optimization (8x faster)

Large JSON imports are now processed in workers:

  • Non-blocking UI during import
  • Progress indicators for large files
  • Validation happens before import (fail fast)

Benchmarks

Operation Before After Improvement
Edit 100 tokens 4.2s 0.8s 5.2x faster
Edit 500 tokens 22.1s 2.1s 10.5x faster
Import 200 tokens 8.5s 1.1s 7.7x faster
Scroll 1000 rows Janky Smooth 60fps

Tested on M1 Mac with Figma Desktop

What’s Next

We’re not done! Coming soon:

  • ⚡ Real-time collaboration without conflicts
  • 📊 Undo/redo for bulk operations
  • 🔍 Faster search and filtering

Try It Now

Update to the latest version in Figma:

  1. Plugins → Manage Plugins
  2. Find “Winden Tokens”
  3. Click “Update”

Let us know how it performs for your design system! Drop your feedback in our GitHub issues.


Performance metrics based on internal testing with real-world design systems. Your mileage may vary based on system complexity and hardware.