Instructions to use epfl-vita/svi-model with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use epfl-vita/svi-model 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("epfl-vita/svi-model", dtype=torch.bfloat16, device_map="cuda") pipe.to("cuda") prompt = "A man with short gray hair plays a red electric guitar." image = load_image( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png" ) output = pipe(image=image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
Native Workflow ?
#14
by ApexArtist - opened
I dont like to use wrappers , will you create a native workflow ?
从目前来看,这并不存在。我看到了youtube上有人分享了关于svi的workflow,并得到了svi团队的认可。但是,我使用那个工作流以后,效果真的糟透了。我发誓,我没有改动工作流的任何内容。github上,svi团队在说明里写道:Profile:SVI-2.0 Unified - 50 Steps Steps:50 Best For:Best quality, dynamic scenes,我不明白他们想传递什么信息。我关闭了Lightning lora,然后我将Steps设置为50,最终我拿到了一段由色素块组成的视频。只能等待svi团队继续完善。
that means no thanks i will stick with painter long video node..