How to inference a specific checkpoint from a model repo that consists of multiple checkpoints?

[wilfrediscoming]

I want to use this model repo: https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE/tree/main

and specifically the inpainting model "Realistic_Vision_V5.1-inpainting.safetensors"

inpaint = StableDiffusionControlNetInpaintPipeline.from_pretrained(
          "SG161222/Realistic_Vision_V5.1_noVAE",
          controlnet = controlnet,
          variant = "fp16",
          use_safetensors = True,
          vae = vae,
          torch_dtype = torch.float16,
          load_safety_checker = False
          ).to("cuda")

how should I update my code?

[sayakpaul]

You should use from_single_file() for that. Like this:

URL = "https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE/blob/main/Realistic_Vision_V5.1-inpainting.safetensors"

inpaint = StableDiffusionControlNetInpaintPipeline.from_single_file(
          URL,
          controlnet = controlnet,
          variant = "fp16",
          use_safetensors = True,
          vae = vae,
          torch_dtype = torch.float16,
          load_safety_checker = False
          ).to("cuda")

[wilfrediscoming]

I got this error. I mainly want to use inference the model repo using the inpaint checkpoint...

ValueError                                Traceback (most recent call last)
[<ipython-input-2-ea11e9861aa5>](https://localhost:8080/#) in <cell line: 46>()
     44 URL = "https://huggingface.co/SG161222/Realistic_Vision_V5.1_noVAE/blob/main/Realistic_Vision_V5.1-inpainting.safetensors"
     45 
---> 46 inpaint = StableDiffusionControlNetInpaintPipeline.from_single_file(
     47           URL,
     48           controlnet = controlnet,

4 frames
[/usr/local/lib/python3.10/dist-packages/accelerate/utils/modeling.py](https://localhost:8080/#) in set_module_tensor_to_device(module, tensor_name, device, value, dtype, fp16_statistics, tied_params_map)
    343     if value is not None:
    344         if old_value.shape != value.shape:
--> 345             raise ValueError(
    346                 f'Trying to set a tensor of shape {value.shape} in "{tensor_name}" (which has shape {old_value.shape}), this look incorrect.'
    347             )

ValueError: Trying to set a tensor of shape torch.Size([320, 9, 3, 3]) in "weight" (which has shape torch.Size([320, 4, 3, 3])), this look incorrect.

[sayakpaul]

We recently fixed it (#6975). Could you ensure you're using latest diffusers installed from the main?

[wilfrediscoming]

I am using the version (0.26.3)

[sayakpaul]

How about uninstall it and then reinstalling with:

pip install git+https://github.com/huggingface/diffusers