XODA-LTXV-Video-Pack
0.9.5 base model update:
TBA version 8 workflows released!
0.9.1 base model update:
Initial Release video:
click show more, full details are below!
Consistent Video Variations can be provided by using my String Painter Node!
A comprehensive workflow pack for ComfyUI implementing the LightTricks LTXV video generation model, complete with example workflows, prompts, and custom tools for streamlined video generation.
Overview
This pack brings together essential workflows, tools, and resources for working with the groundbreaking LTXV real-time video generation AI model from LightTricks. It includes official example workflows, custom prompt generation tools, and extensive prompt lists to help users get started quickly.
Contents
Example Workflows
The pack includes the official ComfyUI maintainer and LTXV author-provided workflows:
AUTHOR-ltxvideo-i2v.json
AUTHOR-ltxvideo-v2v.json
COMFY-ltxv_image_to_video.json
COMFY-ltxv_text_to_video.json
Resources and Tools
Getting Started Guide
Comprehensive setup instructions
Required models and dependencies
Available at: CivitAI Article #9115
LTXV PromptGen LLM Module
Custom LLM module specifically designed for LTXV video generation
Repository: LLM-Base-Prompts/LTXV-PromptGen
Curated Prompt Lists
110 professionally crafted prompts
Organized into 11 categories
Located at: LTXV-PromptGen/LTXV-PromptLists
Integration with DJZ-Nodes
Custom nodes for workflow automation
Includes Zenkai-Prompt nodes for automated prompt handling
Seamless integration with provided prompt lists
Located at: DJZ-Nodes/Prompts
Getting Started
Clone this repository
Follow the setup guide at CivitAI Article #9115
Install required models and dependencies
Load example workflows in ComfyUI
Features
Official LTXV example workflows
Custom prompt generation tools
Extensive prompt library
Automation capabilities
Integration with existing tools
Requirements
Please refer to the Getting Started Guide for a complete list of required models and dependencies.
Description
FAQ
Comments (14)
Thank you! FYI, tried using the PixArt clip instead of T5 and it didn't work. However the Flux t5xxl_fp16 worked perfectly.
I'd love to see a video or a how-to to make the video loop. My use case is making AD&D character and NPC portraits that "live" but it is a bit jarring when they loop abruptly. Lumalabs.ai, for example, has a check-box to loop your videos which I love but I have cancelled my account a long while back. Having that feature on my desktop would be great.
I'm not sure about this persons workflow, but the default workflow I use, outputs to an animating .WEBP image file that automatically loops.
@346542742384560934687 yes, .webp can make a loop, but it will not start where it finished, a true loop is seamless not just playing over and over (with a visible cut) :)
I have other workflows "Looped Motion" which creates looping video with other setups.
@TheKraken2024 Looping with a pingpong (reversing the footage) is an option with saving to MP4 and this is a loop, but it has the same problem. I'll be keeping an eye out for a solid method as i wish to make "LoopedMotion3"
@driftjohnson Thanks and nice chatting with you on your livestream yesterday. I hope you do more of them.
thanks for this info. I was just trying to figure out if I can avoid downloading the two 10gb files each.
@Cavernust iirc if you use the native comfyui based workflows, or ones that use that method, you can omit the pixart text encoders, afaik they are only used by the authors type approach. I included both in the \examples\ folder so you can see both types. Mine are all based on the native comfyui type.
If you don't mind, would you share your process in creating the "pose host" in the youtube video?
As i use this to brand all the tutorials on my channel, it's a part of how i market my material. However I actually gave it away in my first video on the channel, it's shown, but not explained.
All i can say is that i used a comfy workflow to build them, so you can do it all "in workflow".
At a certain subscriber goal i will explain it in more detail - but for now i have to keep people guessing ;)
There is no way to make these videos loop seamlessly?
what you want to do required some planning, you want the final frame to be the first frame, that is the "loop" - however the model will see this as "no motion", so the trick is to chain 2 or more videos and requires control of the last frame, or to reverse the clip. So this is tricky, because you don't want it to be obviously reversed footage as this is easy to spot. Currently there is no "first and last frame" in the standard nodes for this model. People will create custom nodes that might allow this. We can take the last frame from a video and use it to generate a new video - so this is a challenge for workflow builders and custom node devs. The model is 2 days old, so you can solve this with your own workflow or write your own node. (or wait)
@driftjohnson Awesome thank you for your very detailed reply, I appreciate it
@moehawk we might have a soft solution for this in the V2 pack release, but it's an Extended ping pong really, so i cheated, it's called LoopedMotion3. I'm hoping we get a true solution for this, V3 pack will add A Lot of new workflows, been hammering away in my lab with a bunch of cool nodes from other authors
@driftjohnson Seriously you are a legend my friend! I really appreciate you taking your time and working on this