ltx workflow

LTX 2.3 Prompt Generator & ComfyUI Workflow JSON

Match your GPU, generate ComfyUI workflows, and enhance prompts for LTX-2.3 and LTX-2.3 Distilled. Supports FP8 quantized variants — runs on 16GB+ VRAM.

taeltx2_3.safetensorsspatial upscaler x2ComfyUI JSON export16GB–32GB VRAM

VRAM Adapter

Select your GPU VRAM to see compatible LTX-2.3 models:

LTX-2.3 Dev (FP8, Kijai)ltx-2.3-22b-dev_transformer_only_fp8_input_scaled.safetensors16GB VRAM
HuggingFace →
LTX-2.3 Distilled (FP8 v3, Kijai)ltx-2.3-22b-distilled_transformer_only_fp8_input_scaled_v3.safetensors16GB VRAM
HuggingFace →
Spatial Upscaler x2ltx-2.3-spatial-upscaler-x2-1.0.safetensors
HuggingFace →
LTX-2.3 VAEtaeltx2_3.safetensorsRequired
HuggingFace →

LTX 2.3 Prompt Generator

Basic prompt enhancement — sign in for director-level cinematic prompts

ComfyUI Workflow JSON Generator

{
  "1": {
    "class_type": "CheckpointLoaderSimple",
    "inputs": {
      "ckpt_name": "ltx-2.3-22b-dev.safetensors"
    }
  },
  "2": {
    "class_type": "CLIPTextEncode",
    "inputs": {
      "text": "A cinematic shot of a mountain landscape at golden hour",
      "clip": [
        "1",
      
...

LTX-2.3 Model Downloads

LTX-2.3 Dev

Official
ltx-2.3-22b-dev.safetensors

Full dev model. Flexible and trainable. Recommended 32GB+ VRAM.

~42 GB32GB+ VRAM
Download on HuggingFace →

LTX-2.3 Distilled

Recommended
ltx-2.3-22b-distilled.safetensors

Distilled version. 8 steps, CFG=1. Faster inference, same quality.

~42 GB32GB+ VRAM
Download on HuggingFace →

LTX-2.3 Dev (FP8, Kijai)

16GB VRAM
ltx-2.3-22b-dev_transformer_only_fp8_input_scaled.safetensors

FP8 quantized by Kijai. Runs on 16GB VRAM. Requires 40xx+ GPU for fp8 matmuls. Place in models/checkpoints/.

25 GB16GB+ VRAM
Download on HuggingFace →

LTX-2.3 Distilled (FP8 v3, Kijai)

16GB VRAM
ltx-2.3-22b-distilled_transformer_only_fp8_input_scaled_v3.safetensors

FP8 distilled v3 by Kijai. Best for 16GB VRAM. 8 steps, CFG=1.

25 GB16GB+ VRAM
Download on HuggingFace →

Spatial Upscaler x2

ltx-2.3-spatial-upscaler-x2-1.0.safetensors

spatial upscaler x2 for two-stage pipelines. Place in models/latent_upscale_models/.

~1 GB
Download on HuggingFace →

LTX-2.3 VAE

Required
taeltx2_3.safetensors

VAE by Kijai. Required for ComfyUI workflows. Place in models/vae/.

~0.5 GB
Download on HuggingFace →

Official Resources

LTX-2.3 on HuggingFace →

Official model weights by Lightricks

ComfyUI-LTXVideo →

Official ComfyUI nodes & example workflows

Kijai FP8 Models →

FP8 quantized variants for 16GB VRAM

LTX-Video GitHub →

Official LTX-Video model repository

ComfyUI →

Node-based UI for running diffusion models

ComfyUI Manager →

Install LTXVideo nodes via Manager