Mix of realisian, dark sushimix and toonify
recommended settings: sampler: DPM 2 karras or euler a (use DDIM for a more anime look), clip skip: 2
steps: 30
VAE: none, already included.
share your genrations and enjoy!
Description
first version
FAQ
Comments (8)
I think the config yaml is missing. I get errors then trying to load this model
in local or colab? i can't test in local as i don't have a good pc to run S.D, in colab i can use it normally, have you downloaded the model correctly, you can try again to download it, or you can try from tensorart:
Local with webui. But it seems, it doesn't load very well on my laptop (RTX 3070 8 GB with 15 GB memory). I tried it earlier on my Desktop (GTX 1080 with 8 GB VRAM and 32 GB of RAM) and I was able to load it. Took me a loot of research and trial and error to get all the embeddings though.
Is there a know issue with 16 GB RAM systems? I was able to run all other models (20 in total) from civitai on the laptop without any issues
I was getting
Loading weights [82c2e35770] from C:\Projekte\AI\stable-diffusion-webui\models\Stable-diffusion\asianmixW50BackedVAE_v10.safetensors
Creating model from config: C:\Projekte\AI\stable-diffusion-webui\configs\v1-inference.yaml
LatentDiffusion: Running in eps-prediction mode
DiffusionWrapper has 859.52 M params.
Loading VAE weights found near the checkpoint: C:\Projekte\AI\stable-diffusion-webui\models\VAE\asianmixW50BackedVAE_v10.safetensors
changing setting sd_model_checkpoint to asianmixW50BackedVAE_v10.safetensors [82c2e35770]: RuntimeError
Traceback (most recent call last):
File "C:\Projekte\AI\stable-diffusion-webui\modules\shared.py", line 605, in set
self.data_labels[key].onchange()
File "C:\Projekte\AI\stable-diffusion-webui\modules\call_queue.py", line 13, in f
res = func(*args, **kwargs)
File "C:\Projekte\AI\stable-diffusion-webui\webui.py", line 226, in <lambda>
shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: modules.sd_models.reload_model_weights()), call=False)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_models.py", line 552, in reload_model_weights
load_model(checkpoint_info, already_loaded_state_dict=state_dict)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_models.py", line 484, in load_model
load_model_weights(sd_model, checkpoint_info, state_dict, timer)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_models.py", line 336, in load_model_weights
sd_vae.load_vae(model, vae_file, vae_source)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_vae.py", line 143, in load_vae
loadvae_dict(model, vae_dict_1)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_vae.py", line 168, in loadvae_dict
model.first_stage_model.load_state_dict(vae_dict_1)
File "C:\Projekte\AI\stable-diffusion-webui\venv\lib\site-packages\torch\nn\modules\module.py", line 2041, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for AutoencoderKL:
Missing key(s) in state_dict: "encoder.conv_in.weight", "encoder.conv_in.bias",
on my laptop. Using stablediffusion/configs/stable-diffusion/v2-inference.yaml at main · Stability-AI/stablediffusion · GitHub with the name of the model next to it (google results pointed in this direction), but it had other errors about resolution. I changed a few of the values and the model loaded, but then i got
Creating model from config: C:\Projekte\AI\stable-diffusion-webui\models\Stable-diffusion\asianmixW50BackedVAE_v10.yaml
LatentDiffusion: Running in eps-prediction mode
Failed to create model quickly; will retry using slow method.
LatentDiffusion: Running in eps-prediction mode
changing setting sd_model_checkpoint to asianmixW50BackedVAE_v10.safetensors [82c2e35770]: RuntimeError
Traceback (most recent call last):
File "C:\Projekte\AI\stable-diffusion-webui\modules\shared.py", line 605, in set
self.data_labels[key].onchange()
File "C:\Projekte\AI\stable-diffusion-webui\modules\call_queue.py", line 13, in f
res = func(*args, **kwargs)
File "C:\Projekte\AI\stable-diffusion-webui\webui.py", line 226, in <lambda>
shared.opts.onchange("sd_model_checkpoint", wrap_queued_call(lambda: modules.sd_models.reload_model_weights()), call=False)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_models.py", line 552, in reload_model_weights
load_model(checkpoint_info, already_loaded_state_dict=state_dict)
File "C:\Projekte\AI\stable-diffusion-webui\modules\sd_models.py", line 478, in load_model
sd_model = instantiate_from_config(sd_config.model)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\util.py", line 89, in instantiate_from_config
return get_obj_from_str(config["target"])(**config.get("params", dict()))
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\models\diffusion\ddpm.py", line 550, in init
super().__init__(conditioning_key=conditioning_key, args, *kwargs)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\models\diffusion\ddpm.py", line 92, in init
self.model = DiffusionWrapper(unet_config, conditioning_key)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\models\diffusion\ddpm.py", line 1314, in init
self.diffusion_model = instantiate_from_config(diff_model_config)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\util.py", line 89, in instantiate_from_config
return get_obj_from_str(config["target"])(**config.get("params", dict()))
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\diffusionmodules\openaimodel.py", line 721, in init
) if not use_spatial_transformer else SpatialTransformer(
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\attention.py", line 307, in init
[BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d],
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\attention.py", line 307, in <listcomp>
[BasicTransformerBlock(inner_dim, n_heads, d_head, dropout=dropout, context_dim=context_dim[d],
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\attention.py", line 260, in init
self.ff = FeedForward(dim, dropout=dropout, glu=gated_ff)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\attention.py", line 67, in init
) if not glu else GEGLU(dim, inner_dim)
File "C:\Projekte\AI\stable-diffusion-webui\repositories\stable-diffusion-stability-ai\ldm\modules\attention.py", line 52, in init
self.proj = nn.Linear(dim_in, dim_out * 2)
File "C:\Projekte\AI\stable-diffusion-webui\venv\lib\site-packages\torch\nn\modules\linear.py", line 96, in init
self.weight = Parameter(torch.empty((out_features, in_features), **factory_kwargs))
RuntimeError: [enforce fail at ..\c10\core\impl\alloc_cpu.cpp:72] data. DefaultCPUAllocator: not enough memory: you tried to allocate 134217728 bytes.
@TScitai I asked an IA about the error and this is what he answered me: "It looks like there is an error occurring while loading weights for a model. Specifically, the error message is indicating that there are missing keys in the state_dict for the "encoder.conv_in.weight" and "encoder.conv_in.bias". This suggests that there may be an issue with the encoder component of the model.
It's difficult to say exactly what's causing the error without more information about the model and the code that's being used to load the weights. However, one possibility is that the weights file is not compatible with the current version of the model. Another possibility is that there is a bug in the code that's preventing the weights from being loaded correctly.
To diagnose the issue further, it may be helpful to review the code that's being used to load the weights and compare it to the code used to train and save the model. Additionally, it may be useful to try loading the weights into a fresh instance of the model to see if the issue persists."
but even so it seems strange to me because in google colab it works perfectly, and I have read that sometimes S.D is deconfigured out of nowhere, by the way this model is for S.D 1.5. if you use S.D 2 maybe that's why it gives error, it would be a good idea if you could try in another pc if possible or in google colab itself.
@WhiteZ Weird indeed, because on my weaker (but more RAM) Desktop PC it works. I use Automatic1111's webui, Version 1.4 on both, with some extensions. Tried with all extensions disabled same, still same. Will have a deeper look into it.
But anyways was able to reproduce the spider girl on Desktop after getting all the embeddings and loras. Will try to figure out whats causing it
@TScitai a reinstallation of sd can solve the problem, or even a reinstallation of windows, or drivers, it can be many things, but I'm glad it works for you at least on a pc
I love this Checkpoint, please keep it updated!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.