Initial Version V1
Runtime profile
Estimated VRAM requirement
Estimate unavailable
21.4 GB across 3 of 8 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.
Requirements
11 requirementsgemma_3_12B_it_fp4_mixed.safetensors
File unverifiedAI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
The LTX 2.3 12GB GGUF DIRECTOR workflow is for controlling LTX video generations from images, text prompts, and custom audio; it can create and save video.
The Director node is described as a video timeline editor for loading images, placing text prompts, and adding custom audio.
Its 42-node graph includes Director, sampling, decoding, latent upscaling, audio preview, video creation, and video saving nodes.
The description says this workflow replaces t2v, i2v, ia2v, ta2v, and multi keyframe workflows.
Inputs include images, text prompts, and custom audio.
The workflow covers video creation and saving, audio preview, and latent upscaling.
Version v1.0 is published by Urabewe and uses LTXV 2.3 as its base model.
Listed node packs are WhatDreamsCost-ComfyUI, comfyui-kjnodes, and ComfyUI-GGUF.
Listed model files include gemma_3_12B_it_fp4_mixed.safetensors, ltx-2-19b-distilled-lora-384.safetensors, ltx-2-3-22b-dev-Q4_K_M.gguf, ltx-2-spatial-upscaler-x2-1.0.safetensors, ltx-2.3-spatial-upscaler-x2-1.1.safetensors, and ltx-2.3_text_projection_bf16.safetensors.
Other listed model files are ltx-2.3-22b-distilled-1.1_lora-dynamic_fro09_avg_rank_111_bf16.safetensors, LTX-2.3-OmniNFT-RL-Lora_bf16.safetensors, LTX23_audio_vae_bf16.safetensors, LTX23_video_vae_bf16.safetensors, and taeltx23.safetensors.
For timeline-style control, place images, text prompts, and custom audio in the Director node.
Suggestion · not verified
The source describes this as a 12GB GGUF workflow.
Before using it, verify that each listed node pack and model file is available under the exact name shown.
Suggestion · not verified
The Director node is described as a work in progress, and no license is specified for the workflow.
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
3 excerptsSource context: 2291 downloads · Type Workflows · Base model LTXV 2.3
Text encoder · Civitai · 2817877
LoRA · 7.15 GB · Hugging Face · Lightricks/LTX-2
Checkpoint · 13.3 GB · GGUF · Unknown
Upscaler · 950 MB · Hugging Face · Lightricks/LTX-2
ltx-2.3_text_projection_bf16.safetensors
Not resolvedText encoder · Unknown
LTX23_audio_vae_bf16.safetensors
Not resolvedVAE · Hugging Face
LTX23_video_vae_bf16.safetensors
File unverifiedVAE · Hugging Face · LTX23 Video VAE Bf16.safetensors
taeltx23.safetensors
Not resolvedVAE · Unknown
ComfyUI-GGUF
PossibleNode pack · Registry
Node pack · Registry
Node pack · Registry
Source context: 1744 downloads · Type Workflows · Base model LTXV 2.3
Source context: 1739 downloads · Type Workflows · Base model LTXV 2.3