Hugging Face's logo

nakayama
/
DeDeDe

Text-to-Image
Diffusers
StableDiffusionPipeline
stable-diffusion
Inference Endpoints
Model card Files Community
1
DeDeDe / README-en.md
nakayama's picture
nakayama
Update README-en.md
b01307a
preview code
|
raw
history blame contribute delete
11.8 kB

Please read the licence terms below the model card carefully before use.

DeDeDe is a Stable Diffusion Model that has been modified to make animated characters easier to create.
The base model, DreamLike Diffusion 1.0, was difference-merged with Trinart Characters v2 Derrida using Stable Diffusion 1.4.
From there, IN0~5 was adjusted with DreamLike Photoreal 1.0 and finetuned with 30000 images output from SD2.1, Novel AI, WD1.3/1.4, CoolJapan Diffusion 2.1 and Dreamlike Photoreal 2.0.

The following prompts/negative prompts are recommended for use:
P: best quality, masterpiece
NP: 3d, flat shading, flat color, retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb

Examples 

(((best quality, masterpiece, 8k))), detailed anime style of 1girl sitting in room and reading book wearing school uniform and wavy detailed pink hair pink and detailed yellow eye yellow, smiling
Negative prompt: [[3d]], (((flat shading, flat color))), retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb
Steps: 25, Sampler: Euler a, CFG scale: 8, Seed: 2277801742, Size: 512x768, Model hash: 6d1729a039, Denoising strength: 0.75, Clip skip: 2, ENSD: 31337, Hires upscale: 1.5, Hires upscaler: Latent

(((best quality, masterpiece, 8k))), detailed anime style of anime 1girl bust shot sitting and dipping in river and wetty wearing white transparent onepiece dress with detailed wavy pink hair pink and hetailed yellow eye yellow, water splash in gorgeous scene secret garden
Negative prompt: [[[3d]]], (((flat shading, flat color))), retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb
Steps: 20, Sampler: DPM++ SDE Karras, CFG scale: 8, Seed: 1117106368, Size: 512x768, Model hash: 6d1729a039, Denoising strength: 0.7, Clip skip: 2, ENSD: 31337, Hires resize: 768x1152, Hires steps: 5, Hires upscaler: Latent

(((best quality, masterpiece))), detailed anime style of bunny girl bishoujo from front wearing intricate frill jirai kei bikini with detailed wavy pink hair pink and detailed yellow eye yellow and lying on the bed in (((glitter harajuku kawaii messy room with flower and candy)))
Negative prompt: 3d, (((flat shading, flat color))), retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb, eating
Steps: 50, Sampler: DPM++ SDE Karras, CFG scale: 5.5, Seed: 911018641, Size: 768x512, Model hash: 6d1729a039, Denoising strength: 0.2, Clip skip: 2, ENSD: 31337, Hires resize: 1152x768, Hires steps: 10, Hires upscaler: ESRGAN_4x

(((best quality, masterpiece))), detailed anime style of 1girl bishoujo full body standing and looking from viewer and wearing classical frill dress with cape with wavy detailed pink hair pink and detailed yellow hair yellow in scenic view british fantastic landscape with flowing and golden hour, dynamic pose
Negative prompt: [[[[3d]]]], (((flat shading, flat color))), retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb
Steps: 15, Sampler: DPM++ SDE Karras, CFG scale: 5.5, Seed: 2149903685, Size: 768x512, Model hash: 6d1729a039, Denoising strength: 0.75, Clip skip: 2, ENSD: 31337, Hires resize: 1152x768, Hires steps: 10, Hires upscaler: Latent

(((best quality, masterpiece, 8k))), detailed anime style of 1boy cowboy shot wearing samurai outfit with flat chest and fighting pose, fist, motion blur
Negative prompt: 3d, flat shading, flat color, retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb
Steps: 20, Sampler: Euler a, CFG scale: 8, Seed: 2136917906, Size: 512x768, Model hash: 6d1729a039, Denoising strength: 0.6, Clip skip: 2, ENSD: 31337, Hires resize: 768x1152, Hires steps: 20, Hires upscaler: Latent

(((best quality, masterpiece, 8k))), detailed photorealistic style of 1boy cowboy shot wearing mad max outfit with flat chest and fighting pose, fist, motion blur, abandoned city background
Negative prompt: (((flat shading, flat color))), retro style, 1980s, 1990s, 2000s, 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019, lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, artist name, inaccurate limb
Steps: 20, Sampler: Euler a, CFG scale: 8, Seed: 3787706619, Size: 768x512, Model hash: 6d1729a039, Denoising strength: 0.7, Clip skip: 2, ENSD: 31337, Hires resize: 1152x768, Hires steps: 5, Hires upscaler: Latent

Merge/Learning procedure

The symbolic sequence listed next to the model is the hash value of the ckpt listed at the time of model selection at Automatic1111 Webui Commit Hash c98cb0f8ecc904666f47684e238dd022039ca16f.

  1. Added Trinart Derrida and Stable Diffusion 1.4 diff to Dreamlike Diffusion 1.0.

    Interpolation Method Primary Model Secondary Model Tertiary Model Merge Name
    Add Difference @ 1.0 DreamLike Diffusion 1.0(0aecbcfa2c) TrinArt Characters v2 Derrida(42d3f359b0) Stable Diffusion 1.4(fe4efff1e1) DDD_pre1(d1ac03017b)

  2. Editing IN00~IN05 in the hierarchy of Unet in DDD_pre1 using Dreamlike Photoreal 1.0 with hierarchical merging.

    Model: A Model: B Weight Base alpha Merge Name
    DDD_pre1(d1ac03017b) Dreamlike Photoreal 1.0(f403e4e2a5) 0.45,0.45,0.4,0.35,0.3,0.25,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0 0 DDD_pre2(601ec74593)

  3. Fine-tuned DDD_pre2 using material images. The material images are output from other Stable Diffusion models.
    The services/models used in the preparation were SD2.1, Novel AI, WD1.3/1.4, CoolJapan Diffusion 2.1, and Dreamlike Photoreal 2.0.
    The total number of images is about 30,000. These were combined with flipped images and trained for 60000 steps at a learning rate of 5e-6.
    The model generated by this process is DDD_pre3 (4709475652).

  4. Merged DDD_pre2 to DDD_pre3 with weighted sum.

    Interpolation Method Primary Model Secondary Model Merge Name
    Weighted Sum @ 0.5 DDD_pre3(4709475652) DDD_pre2(601ec74593) DeDeDe(6d1729a039)

Flavours

DeDeDe_ip2p_0.7_0.8.ckpt/DeDeDe_ip2p_0.7_1.0.ckpt

The task vectors are extracted from the Instruct pix2pix model and added together.
The sizes are DeDeDe 0.8/Instruct Pix2Pix 0.7 and DeDeDe 1.0/Instruct Pix2Pix 0.7.
The following licence is inherited from Instruct Pix2Pix.
Copyright (c) 2023 Ren Nakayama
Released under the MIT Licence
https://huggingface.co/nakayama/DeDeDe/blob/main/MIT-License

DeDeDe_controlnet_*.pth/DeDeDe_webui_controlnet_*.safetensors

This is a merge of DeDeDe with the model used in ControlNet.
See here for specific instructions.

DeDeDeP

https://huggingface.co/nakayama/DeDeDeP
Based on DeDeDe and edited with Dreamlike Photoreal 1.0 and Hierarchical Merging.
The output is more realistic than DeDeDe.

Notes

When uploading your output to social networking sites etc. it would be great if you could add tags or other features such as #DeDeDeArt.
I will be there to see it.

License

This model is licesed under a modified CreativeML OpenRAIL-M license.

  • You can't host or use the model or its derivatives on websites/apps/etc., from which you earn, will earn, or plan to earn revenue or donations. If you want to, please email us at contact@dreamlike.art
  • You are free to host the model card and files (Without any actual inference or finetuning) on both commercial and non-commercial websites/apps/etc. Please state the full model name (Dreamlike Diffusion 1.0 / Dreamlike Photoreal 1.0) and include a link to the model card ( https://huggingface.co/dreamlike-art/dreamlike-diffusion-1.0 / https://huggingface.co/dreamlike-art/dreamlike-photoreal-1.0/ )
  • You are free to host the model or its derivatives on completely non-commercial websites/apps/etc (Meaning you are not getting ANY revenue or donations). Please state the full model name (Dreamlike Diffusion 1.0 / Dreamlike Photoreal 1.0) and include a link to the model card ( https://huggingface.co/dreamlike-art/dreamlike-diffusion-1.0 / https://huggingface.co/dreamlike-art/dreamlike-photoreal-1.0/ )
  • You are free to use the outputs of the model or the outputs of the model's derivatives for commercial purposes in teams of 10 or less
  • You can't use the model to deliberately produce nor share illegal or harmful outputs or content
  • The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
  • You may re-distribute the weights. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the modified CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully) Please read the full license here: https://huggingface.co/nakayama/DeDeDe/blob/main/License.md