[Bug] Error(s) in loading state_dict for AutoencoderKL #619
Description
Checklist
- 1. I have searched for related issues and FAQs (https://github.com/mit-han-lab/nunchaku/blob/main/docs/faq.md) but was unable to find a solution.
- 2. The issue persists in the latest version.
- 3. Please note that without environment information and a minimal reproducible example, it will be difficult for us to reproduce and address the issue, which may delay our response.
- 4. If your report is a question rather than a bug, please submit it as a discussion at https://github.com/mit-han-lab/ComfyUI-nunchaku/discussions/new/choose. Otherwise, this issue will be closed.
- 5. I will do my best to describe the issue in English.
Describe the Bug
When I use Nunkaku's official Qwen Image workflow, there is an error when running 'load VAE'.
Qwen Image Edit also has the same issue
Error(s) in loading state_dict for AutoencoderKL:
size mismatch for encoder.conv_in.weight: copying a param with shape torch.Size([96, 3, 3, 3, 3]) from checkpoint, the shape in current model is torch.Size([128, 3, 3, 3]).
size mismatch for encoder.conv_in.bias: copying a param with shape torch.Size([96]) from checkpoint, the shape in current model is torch.Size([128]).
size mismatch for encoder.conv_out.weight: copying a param with shape torch.Size([32, 384, 3, 3, 3]) from checkpoint, the shape in current model is torch.Size([32, 512, 3, 3]).
size mismatch for decoder.conv_in.weight: copying a param with shape torch.Size([384, 16, 3, 3, 3]) from checkpoint, the shape in current model is torch.Size([512, 16, 3, 3]).
size mismatch for decoder.conv_in.bias: copying a param with shape torch.Size([384]) from checkpoint, the shape in current model is torch.Size([512]).
size mismatch for decoder.conv_out.weight: copying a param with shape torch.Size([3, 96, 3, 3, 3]) from checkpoint, the shape in current model is torch.Size([3, 128, 3, 3]).
size mismatch for quant_conv.weight: copying a param with shape torch.Size([32, 32, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([32, 32, 1, 1]).
size mismatch for post_quant_conv.weight: copying a param with shape torch.Size([16, 16, 1, 1, 1]) from checkpoint, the shape in current model is torch.Size([16, 16, 1, 1]).
Environment
Windows 10
Python 3.12.9
CUDA 12.6
pytorch 2.7.1+126
comfyui：0.3.60
Reproduction Steps
workflow:
- https://github.com/nunchaku-tech/ComfyUI-nunchaku/blob/main/example_workflows/nunchaku-qwen-image.json
- https://github.com/nunchaku-tech/ComfyUI-nunchaku/blob/main/example_workflows/nunchaku-qwen-image-edit-2509.json
vae:
clip:
model: