LTX 2.5 GGUF

LTX 2.5 is an open-weights video generation model from Lightricks. This repository ships GGUF quants of the 22B distilled transformer. It generates video with synchronized audio from text, a single frame, or first and last frames.

Hugging Face repositoryLTX-2.5-GGUF on Hugging Face

Example output

Video from CivitAI

Available Quantizations

Quant	Size	File
Q4_0	11.65 GB	ltx-2.5-22b-distilled-transformer-Q4_0.gguf
Q8_0	21.22 GB	ltx-2.5-22b-distilled-transformer-Q8_0.gguf
Q8_CR	20.04 GB	ltx-2.5-22b-distilled-transformer-Q8_CR.gguf

Model Input and Output

Variants and inputs

Variant	Inputs
Text-to-Video	A text prompt only.
Image-to-Video	A text prompt and one source frame.
First and Last Frame to Video	A text prompt and two keyframe images.

Outputs

Field	Value
Audio	Synchronized stereo audio
Frame rate	24 FPS
Resolution	Up to 2K with the spatial upscaler

ComfyUI Setup

Load the model with the GGUF loader node from comfyui-gguf-reboot. The loader node gives the best performance.

Dependencies

Place these files in the ComfyUI model folders before running a workflow.

Component	File	Folder
Text encoder	Gemma 4 12B with projection	models/clip/ or text_encoders/
Video VAE	ltx-2.5-video-vae-bf16.safetensors	models/vae/
Audio VAE	ltx-2.5-audio-vae-bf16.safetensors	models/vae/
Spatial upscaler	ltx-2.5-latent-spatial-upscaler-x2-bf16-1.0.safetensors	models/upscale_models/

Workflow downloads

Import these JSON files into ComfyUI.

JSON fileText-to-Video workflowJSON fileImage-to-Video workflowJSON fileFirst and Last Frame to Video workflow