Forum Replies Created
-
Posts
-
The extension has been updated and doesn’t work with the original models.
The lessons have been updated. In essence, download the LoRAs from the following link.
https://huggingface.co/conrevo/AnimateDiff-A1111/tree/main/lora_v2
They only work with this v2 motion model.
https://huggingface.co/conrevo/AnimateDiff-A1111/blob/main/motion_module/mm_sd15_v2.safetensors
Hi, the flux sampler is there if you run the notebook with the default version. I guess you must have removed the version hash so it is using the latest version.
The flux sampler was removed yesterday. https://github.com/lllyasviel/stable-diffusion-webui-forge/issues/2554
I now pinned the notebook to a newer version of forge but still have the flux sampler.
Got you. You can make a comfyui workflow to automate it in some degree.
For full automation, it is better to write a custom pipeline.
Hello,
Lessons for changing clothes:
Lessons for changing background:
Let me know how it goes.
I just tested the notebook with the sample training images. It is working correctly. Is this something related to your training image set?
I haven’t trained any but kohya supports training SD 3.5 in the sd3 branch.
It is a known bug in Forge: https://github.com/lllyasviel/stable-diffusion-webui-forge/issues/2356