You can use my model directly at : Tensort.Art | Mage.Space
_____
Units can contact me if they want to release for commercial or exclusive use ( Facebook | Zalo | Mail )
_____
If you find it useful, please vote 5 stars and post an image to help me, it's really important, I thank you all very much.
_____
📖 Checkpoint Info
Info : This is a model I really like, it can create a lot of great styles when combining anime, sketch, realistic ... styles together. You can use a combination of style art supported by SDXL, see more here. Or I have some wildcards to support, you can refer to below. You can add the keyword “niji style” to increase style sensitivity.
____
You can refer to some of the indicators below to achieve the best image quality :
Steps : > 50
CFG : 9-10
Size : 768x1152 px ( or 800x1200px ), 1024x1024
You can also use hiresfix ( hiresfix is not really good at SDXL, if you use it please consider denoising streng 0.2-0.3 ) or After Detailer
Sampling : Euler a or DPM ++ SDE Karass
📝 Quality Prompt :
Base quality prompt :
(masterpiece,best quality,niji style)Key :
niji style , ghibli, anime style, sketch, realistic, acrylic, gradient background, ...📝 Negative Prompt :
low quality,illustration, 3d, 2d, painting, cartoons, sketch📝 Full album ( 3500 image - prompt) :
_____
Also you can test those prompt tags quickly with my model here : Tensor.art
I have written a colab site that integrates all tools for you to use stablediffusion without configuring your computer, you can refer to : Colab SDVN
I have provided an album with more than 2000 original photos with full prompt, you can refer to understand how to use SDXL : SDVN-SDXLPrompt Kit
In addition, some of my models are available on the Mage.space platform, you can refer to: SDVN Mage
_____
On top of that, SDXL is only in its early stages, I may have many shortcomings. I hope you can give me suggestions for corrections in the next updates. If you find it useful, please vote 5 stars and post an image to help me, it's really important, I thank you all very much
_____
👨🏻💻 Creator : stablediffusion.vn
📢 Discord : Discord SDVN
📢 Facebook : Group SDVN
☕️ Donate me : Bank | Buymeacoffee
Description
The official version
FAQ
Comments (23)
Is the VAE included in the model? if not, what VAE do I use?
Ever find an answer to this?
Very unstable and non predictable. Maybe should wait for next version.
Thank you for your feedback, I will try to improve in the next updated versions
What do you call "Very unstable and unpredictable"? Do you have an example to show me if you have time?
For my part, I have no problem with it, hence my question.
I wish you could train this on 1.5.. is it possible? ;-;
I also want
me too
The model has been placed in the correct folder and is also displayed in the model list, but it cannot be selected. After reading for a while, the result still returns to the previously selected model. What is the situation? Console error : .......size mismatch for model.diffusion_model.output_blocks.7.0.skip_connection.bias: copying a param with shape torch.Size([320]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for model.diffusion_model.output_blocks.8.0.in_layers.0.weight: copying a param with shape torch.Size([640]) from checkpoint, the shape in current model is torch.Size([960])..........
You might have solved this problem by updating SD kernel version to 1.5.1 or later, I just solved this. Replied in case someone else runs into the same issue.
If you generate 256x256 or even 512x512 its just noisy and terrible - anyone knows why?
because sdxl trained on 1024x1024, less than 768 is not supported
@OOOOONNDC Oh I see. It takes way too long to generate 1024 images T_T. 40+ seconds on a A4000
@starwarrior516 40+ seconds is not that much.
@starwarrior516 Well, I am using my iPad Pro M1 and it takes 15+ min to generate 1 image... So 40+seconds is ....
@starwarrior516 Takes me like 11 days to do one by hand on my 10 year old sketch tablet, sooo
Is there a model just like this but with SD1.5 or whatever is not XL?
Can you also make niji v6 style?
wonderful, one of my favorite checkpoints!
I like the style it gives to my images. When I test a prompt, it goes through a lot of very different checkpoints including this one ;-)
thank you for your work
Excellent model! I use it in Fooocus. Very Thanks!!!
I have 2 models that are performing well for character generation. This and niji flat. They're both landing characters well, but backgrounds are kinda lackluster. Need more scene context.
This model is a banger. Quite unique and beautiful results. Thanks for sharing it and making it free to use.
Details
Files
sdvn7Nijistylexl_v1.safetensors
Mirrors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
SDVN7-NijiStyleXL-v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
sdvn7Nijistylexl_v1.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.