LTX2.3 GGUF LOW VRAM VIDEO GENERATION I2V / T2V

A simple, color-coded GGUF workflow for running LTXV-2 in ComfyUI.
Designed to be straightforward and lightweight.

• Tested Hardware
  Most videos: RTX 3060 (12GB VRAM) + 48GB RAM
  Works on RTX 3050 Laptop (4GB VRAM) + 16GB RAM (enable Triton)

Common Issues

Error: LTXVEmptyLatentAudio – AttributeError

Update Kijai Node Pack to v1.2.8

Checkerboard / patterned noise output

Update all nodes + ComfyUI
(Outdated GGUF loader or ComfyUI build)

Required Node Packs

ComfyUI-GGUF
https://github.com/city96/ComfyUI-GGUF

ComfyUI-KJNodes
https://github.com/kijai/ComfyUI-KJNodes (install & update it to nightly)

REQUIRED Models LTX2.3 (only workflow V2)

Diffusion Model → diffusion_models/

https://huggingface.co/unsloth/LTX-2.3-GGUF

VRAMFileUnder 12 GBltx-2.3-22b-dev-Q3_K_M.gguf12 GB+ltx-2.3-22b-dev-UD-Q5_K_M.gguf (recommended)

VAE (both required) → vae/

https://huggingface.co/unsloth/LTX-2.3-GGUF/tree/main/vae

• ltx-2.3-22b-dev_video_vae.safetensors

• ltx-2.3-22b-dev_audio_vae.safetensors

Text Encoders (both required) → text_encoders/

https://huggingface.co/unsloth/LTX-2.3-GGUF/tree/main/text_encoders

• ltx-2.3-22b-dev_embeddings_connectors.safetensors

• gemma_3_12B_it.safetensors

LoRA → loras/

https://huggingface.co/Lightricks/LTX-2.3

• ltx-2.3-22b-distilled-lora-384.safetensors

Spatial Upscaler → latent_upscale_models/

https://huggingface.co/Lightricks/LTX-2.3

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

REQUIRED Models LTX2 (OLD)

Main Directory:

https://huggingface.co/Kijai/LTXV2_comfy/tree/main/

You need to download following models:

Diffusion (dev Q4_k_m less vram Q8_0 more vram):
https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-dev_Q4_K_M.gguf

https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/diffusion_models/ltx-2-19b-dev_Q8_0.gguf

Text Encoders (Gemma3 smaller files less vram, connector) Both are required:
https://huggingface.co/Comfy-Org/ltx-2/tree/main/split_files/text_encoders
https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/text_encoders/ltx-2-19b-embeddings_connector_dev_bf16.safetensors

VAE (bf16 audio + video) Both are required: https://huggingface.co/Kijai/LTXV2_comfy/tree/main/VAE

Latent Upscaler: https://huggingface.co/Lightricks/LTX-2/resolve/main/ltx-2-spatial-upscaler-x2-1.0.safetensors

LoRAs
Distilled: https://huggingface.co/Kijai/LTXV2_comfy/resolve/main/loras/ltx-2-19b-distilled-lora_resized_dynamic_fro09_avg_rank_175_fp8.safetensors

(Optional)
Camera: https://huggingface.co/Lightricks/LTX-2-19b-LoRA-Camera-Control-Dolly-Left/tree/main