Closed
Description
Describe the bug
According to https://github.com/huggingface/diffusers/releases/tag/v0.19.2, i can load Lora by
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16)
pipe.load_lora_weights("stabilityai/stable-diffusion-xl-base-1.0", weight_name="sd_xl_offset_example-lora_1.0.safetensors")
pipe.to("cuda")
but in #4287
you wrote
from diffusers import DiffusionPipeline
import torch
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", torch_dtype=torch.float16)
pipe.load_lora_weights("./sd_xl_offset_example-lora_1.0.safetensors")
Are these identical?
pipe.load_lora_weights("stabilityai/stable-diffusion-xl-base-1.0", weight_name="sd_xl_offset_example-lora_1.0.safetensors")
pipe.to("cuda")
and
pipe.load_lora_weights("./sd_xl_offset_example-lora_1.0.safetensors")
Does that mean i can also load any civitai lora model? Do i need to download and put this in local path? Can i reference it by url and load it from there similar to huggingface model_id? How do you suggest i should load lora weight from civitai? i.e https://civitai.com/models/112902/dreamshaper-xl10
Reproduction
N/A
Logs
No response
System Info
ec2
Who can help?
No response