This is a character lora trained for the hunyuan text-to-video generation model
Enjoy summoning Makima to do anything you want them to do for you...
Triggerword: ```kxsr```
suggested prompting format: "kxsr, a lively kxsr pushes her shopping cart violently down the aisle"
Until support for lora loading is on the main branch for Kijai's HunyuanVideoWrappper
the only way to load this lora is to do this:
There's a PR submitted for the changes to the hunyuan loader nodes by Kijai: https://github.com/kijai/ComfyUI-HunyuanVideoWrapper/pull/72
Go grab the nodes.py file from that commit and replace your nodes.py in your ComfyUI-HunyuanVideoWrapper folder and then restart your server.
Description
FAQ
Comments (13)
why I can't connect HunyuanVideo Model Loader Node to LoraLoadModelOnly Node
There's a PR submitted for the changes to the hunyuan loader nodes by Kijai: https://github.com/kijai/ComfyUI-HunyuanVideoWrapper/pull/72
Go grab the nodes.py file from that commit and replace your nodes.py in your ComfyUI-HunyuanVideoWrapper folder and then restart your server.
Update the wrapper
How did you create this lora?
Here's a Tutorial for setting it up!
https://civitai.com/articles/9798/training-a-lora-for-hunyuan-video-on-windows
i cant connect loadlora nor loadloramodelonly node to hunyuan pipeline,how did you do that
There's a PR submitted for the changes to the hunyuan loader nodes by Kijai: https://github.com/kijai/ComfyUI-HunyuanVideoWrapper/pull/72
Go grab the nodes.py file from that commit and replace your nodes.py in your ComfyUI-HunyuanVideoWrapper folder and then restart your server.
@Kytra thanks for you and your help,!
Lora not working for me
workflow image : https://imgur.com/a/hGpTRah
Sorry, I have a question, I noticed in your workflow that you can see the preview in HunyuanVideo sampler, is it right? I have the last wrapper running perfect on an RTX3090 but I can't see it the preview during generation.
@argentinox goto comfyui manager and set preview method to auto
Make loras for hunyuan post here:
https://github.com/kijai/ComfyUI-HunyuanVideoWrapper/issues/136
How about an updated workflow to use the new core video nodes instead of the legacy ones? All the HunyuanVideo nodes become obsolete last week when ComfyUI was updated.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.