ltx workflow

Changelog

Updates to ltx workflow and LTX-2.3 model releases.

2026-04-08

Site Launch

  • +Launched ltxworkflow.com with LTX-2.3 support
  • +ComfyUI workflow JSON generator with official node names
  • +VRAM Adapter for 16GB / 24GB / 32GB GPUs
  • +AI prompt enhancer (gpt-5.2 via yunwu.ai)
  • +Model downloads: dev, distilled, FP8 (Kijai), VAE, spatial upscaler
  • +Google OAuth + email/password authentication
  • +Cloud workflow save via Supabase
  • +Setup guide, model page, workflow templates
2026-04-01

LTX-2.3 Released by Lightricks

  • +Official LTX-2.3 (22B parameters) released on HuggingFace
  • +Distilled variant: 8 steps, CFG=1, same quality as dev
  • +FP8 quantized variants by Kijai for 16GB VRAM (requires RTX 40xx+)
  • +New ComfyUI nodes: LTXVConditioning, LTXVScheduler, EmptyLTXVLatentVideo
  • +Spatial upscaler x2 for two-stage pipelines
  • +TAE VAE (taeltx2_3.safetensors) required for all workflows