ComfyUI ltxv model image generation video tutorial

This article brings you a detailed explanation of the work tutorials on using LTXV in ComfyUI, model loading and workflow.

1. Enter the ComfyUI interface, click Workflow to select Browse Templates

Click video to select LTXV lmage to Video

Enter the LTXV lmage to Video interface

2. The clip model needs to load t5xxl_fp16.safetensors, and select ltxv in type

Click on manager

Select model manager

Search t5xxl to load comfyanonymous/flux_text_encoders - t5xxl (fp16)

3. Load Checkpoint requires LTX-Video 2B v0.9 Checkpoint

Click the manager to select model manager to search LTXV load LTX-Video 2B v0.9 Checkpoint

4. Upload the static image you need to change the image

Case picture:

5. Enter text and negative conditions

for example:

Text: Blink your eyes, your hair flutters slightly

Negative conditions: low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly

6. Set width and height in LTXVlmgtovideo. Since your graphics card and memory are relatively small, the width and height are relatively low. If your device is configured with a high device, you can turn the higher point, which is better.

7. Set other settings as default, then click Queue and wait for a few minutes to generate dynamic video.