Instructions to use Alissonerdx/BFS-Best-Face-Swap-Video with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Alissonerdx/BFS-Best-Face-Swap-Video with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Lightricks/LTX-2.3", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("Alissonerdx/BFS-Best-Face-Swap-Video") prompt = "A man with short gray hair plays a red electric guitar." input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png") image = pipe(image=input_image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
models names mismatch + missing
Hi,
Just a quick note to say that few models are incorrectly named in the workflow and therefore mismatch with models downloaded by the .sh.
example with workflow calling the LTX embeddings "ltx-2-19b-embeddings_connector_bf16.safetensors" which is now named "ltx-2-19b-embeddings_connector_distill_bf16.safetensors". Same with the headswap first frame sub-module which has mismatching lora name and the power lora loader which is loading an unknown lora (head_swap_v5 [...]) which is nowhere to be found.
Those issues make the whole workflow extremely hard to use.
regards
Thanks for letting me know, I made some updates and I hope it's closer now. This happened because I already had the script ready for another version that I was going to post, but in the end I didn't.