Model fails to load on AMD GPU
#17
by
SkyyySi
- opened
I'm trying to use this model in Automatic1111's WebUI, but it fails to load each time. Animagine 3.0 or any other model does not have this issue.
My setup:
- Stable Diffusion WebUI commit: https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/1c0a0c4c26f78c32095ebc7f8af82f5c04fca8c0
- Python: 3.11.7
- Torch: 2.3.0+rocm5.7 (latest nightly build for ROCm 6.1)
- All PIP packages were updated to the latest version as of today (8th of June 2024)
- GPU: AMD Radeon RX 7900 XTX
- OS: Arch Linux (fully updated)
Error traceback:
changing setting sd_model_checkpoint to animagine-xl-3.1.safetensors [e3c47aedb0]: RuntimeError
Traceback (most recent call last):
File "/stable-diffusion-webui/modules/options.py", line 165, in set
option.onchange()
File "/stable-diffusion-webui/modules/call_queue.py", line 13, in f
res = func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/stable-diffusion-webui/modules/initialize_util.py", line 181, in <lambda>
shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: sd_models.reload_model_weights()), call=False)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/stable-diffusion-webui/modules/sd_models.py", line 860, in reload_model_weights
sd_model = reuse_model_from_already_loaded(sd_model, checkpoint_info, timer)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/stable-diffusion-webui/modules/sd_models.py", line 826, in reuse_model_from_already_loaded
load_model(checkpoint_info)
File "/stable-diffusion-webui/modules/sd_models.py", line 748, in load_model
load_model_weights(sd_model, checkpoint_info, state_dict, timer)
File "/stable-diffusion-webui/modules/sd_models.py", line 474, in load_model_weights
sd_vae.load_vae(model, vae_file, vae_source)
File "/stable-diffusion-webui/modules/sd_vae.py", line 212, in load_vae
_load_vae_dict(model, vae_dict_1)
File "/stable-diffusion-webui/modules/sd_vae.py", line 239, in _load_vae_dict
model.first_stage_model.load_state_dict(vae_dict_1)
File "/stable-diffusion-webui/modules/sd_disable_initialization.py", line 223, in <lambda>
module_load_state_dict = self.replace(torch.nn.Module, 'load_state_dict', lambda *args, **kwargs: load_state_dict(module_load_state_dict, *args, **kwargs))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/stable-diffusion-webui/modules/sd_disable_initialization.py", line 221, in load_state_dict
original(module, state_dict, strict=strict)
File "/stable-diffusion-webui/venv/lib/python3.11/site-packages/torch/nn/modules/module.py", line 2214, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for AutoencoderKLInferenceWrapper:
Missing key(s) in state_dict: [This part here is a list of what appear to be tokens / model embeddings, which I omitted due to it being extemely long]
skibidi fortnite