Run Locally
MOHAWK_v18VAEBaked.safetensors
6.46 GB · SafeTensor · fp16 · full
Downloads are served by Civitai. Some creators ask for a Civitai login first.
About This Model
STABLE DIFFUSION XL 1.0 · Base Models
Mohawk V1.8 Baked by Aurety is a checkpoint merge aimed at character design, but can also handle a wide range of content. It may pair well with comics loras
Creator notes
MOHAWK : Character_Designer ( Release )
Suggested to me for the time I spend on these models and if you like what I do here, please considerate to donate on this Ko-Fi page :
Follow me on X / Twitter to watch the updates :
________________________________________________________________________________________________
MOHAWK V2.0 Update!
For the first time ( suggested by some members here and students, I opened a KO-FI page ( links above ) .. For those who wish, you can participate by helping me acquire a more powerful GPU than my 2080ti, you never know :) In any case, it would clearly help me work faster. All the best!
SO..
It's a massive upgrade! I had so much catching up to do with my work on the CHEYENNE model.
I've continued to improve realism and the notion of concept art for rich and varied sketched renderings. There will always be time to improve the details in text2image if necessary. In short, CFG is important and has a huge influence on renderings.
In detail, I worked on :
- Realism ++++
- Widening the shot (wide angle) +++
- Architecture +++
- Details in general ++
- textures ++
- ..
It's now a solid base and I'm going to be able to go back a bit to the 2.5D/3D mix specific to Mohawk.
Like Faeia in her great model Faeia said :
However, I am very opposed to having my model and resources I personally train and test and spend hundreds of hours on, be put up on third-party generation sites for their monetary gain where I receive nothing in return.
V1.8 et V1.8(VAE-BAKED) UPDATE //
Hello everyone and peace on earth... I'd like to offer you this new update, which I hope will win you over as it did me. Over 50 merges from Loras and Modeles (too many to mention but I'd like to thank them all for sharing! ).
I've tried to make up for a few lacks in detail, realism, abstraction, text management, sharpness and above all creativity. I think I'm on the right track now!
For the VAE backed version, it's my first one so tell me is something wrong !
This is a major update for the future of MoHawk and I hope you'll download it and, above all, send me your images so that I can better appreciate where I can improve it, thanks to your criticism and feedback.
I wish you all joy and happiness in these very complicated times...
Best regards
Antoine Aurety
___________________________________________________________________________________
v1.6 Update It's a massive update! I encourage you to download it.
First of all, I'd like to thank for his gigantic work on his SDXL model review, where he placed MOHAWK_ among his favourites. This has given MOHAWK a bit of visibility amongst all the wonderful models available here, of which MOHAWK is only the offspring because it's a Merged Model... As well as all the authors of the LORAs who are helping me to take small steps towards this new update. Thank you all so much!
IMPORTANT: This model requires a VAE which is not included. I'm using the classic SDXL.VAE. Alternative VAE :
I've pushed the cursors a bit in all directions (realism, beautiful faces, textures, materials, complex compositions, details, illustration, matte painting, concept art, etc.) and I hope you like the result as much as I do. I try to respect an editorial process where each evolution doesn't erase the previous model, but evolves it in small, sensitive steps.
Thank you to everyone who posts images and gives me feedback and criticism (even bad criticism!). This model was conceived as a support for my 3D courses, and I'm doing it as a hobbyist, with the desire to share my knowledge and my vision. I just hope it reaches you.
Kisses!
______________________________________________________________________________________
v1.4 Update _ More than 30 mergers ( Lora + Checkpoints )... just a few small details but the result is there, I think! It's a nice update that respects my editorial wishes... The emphasis has once again been placed on matte painting, details and a little more contrast... It's reduced the quality of the full-body faces a little, but I've really gained in close-up and detail of the skin and textures in general... It's also a little more realistic.
I hope you like it, and thanks to those who posted their renderings in the gallery. It's really nice to see the model being used!
Kisses
____________________________________________________________________
V1.2 Update : This is an important update as I'm introducing a bit of Landscape in it and it's a part I want to develop for my matte painting courses.
Otherwise, I've refined the human models: the women are more feminine and the men less muscular. They're not "ideal", but I want to be able to have the occasional "broken face", so I'm fine with that. I increased the quality over all..
If you use the "COMICS" add-on, you'll see that the features are more subtle and expressive.
Anyway, I'm making good progress and hope to see some of your renderings from time to time.
Please follow me and download the model !
R1 Update : This version has been improved in all areas (sharpness, shadow, details, textures, composition, etc.). On the whole, you'll get the same images as the Beta version, but with an obvious gain in quality.
I recommend that you upgrade or get this version!
Many thanks.
First of all, I'd like to thank all the model creators on the Civitai community. All credit goes to them.
MOHAWK has been developed with a view to producing cartoonish or realistic 3D character concepts... Successive merges have tried to support this or that need. It's not perfect, but the results are starting to suit me, so I'd like to share it with you and see what you can do with it, and thanks to your criticism, how I can improve it...
MOHAWK is a merge from various excellents models so it can perform too on any subject, not only character design.. Just for reminder.
How do I use it? :
Once loaded, simply describe the character you want in the prompt and use the various styles provided to render Charadesign an
SHA256
4EF5D3ABCD6426E3DB5B8B11495FC921E436EA93E2A0FC07C5F573330B74D6A2
ModelTrends.ai Model ID
#8347
Flag content
Example renders
Prompts and settings shared by the people who made these renders on Civitai. Pick a render to see how it was made.
Settings
Sampler
DPM++ 2M SDE Karras
Steps
39
Guidance
5
Seed
3893145761
Size
512x768
Prompt
berserk, 1girl, solo, dark warrior, long hair, lunar eclipse, blood, destruction, death flowers, armor, Victoria_Frances style AND Kentaro Miura Style, masterpiece, best quality digital artwork, illustrative, highly detailed, yperdetailed, intricate, elegant, super highly detailed, professional digital photo, art station, concept art, 8k, ((devastated, cruel, wounded, agonizing1.39)), ((sitting on a mountain of corpses)), <lora:Victoria_Frances_v2_XL:1>, <lora:Kentaro Miura Style:0,8>
Negative prompt
lowres, worst quality, blurry,
Settings
Sampler
DPM++ 2M SDE Karras
Steps
39
Guidance
5
Seed
1190935470
Size
1024x1536
Prompt
berserk, one girl, 1girl, dark warrior, apocalipse, Eclipse 1.37, Cloak, Armor, Shouting, dark, realistic, kentaro miura style, AND dark and ghostly, edvard munch masterpiece, best quality digital artwork, illustrative, highly detailed, surreal painting, epic surrealism 8k oil painting, <lora:Kentaro Miura Style:0.5>, <lora:FF-Style-Edvard-Munch-vpred:1>
Negative prompt
lowres, worst quality, blurry,
Settings
Sampler
DPM++ 2M Karras
Steps
40
Guidance
7
Seed
723192733
Size
888x1176
Prompt
(best quality, high quality:1),(homer simpson:0.2),
ohwx man as a lego figure,(lego style),3d,realistic,white shirt,blue pants,lego city street,
<lora:homer_simpson_sdxl_v1:1.2>
Negative prompt
worst quality,low quality,logo,watermark,signature,glasses,fingers
Settings
Sampler
DPM++ 2M SDE Karras
Steps
39
Guidance
5
Seed
3892627800
Size
540x675
Prompt
berserk, one man, dark warrior , ((Eclipse)), Cloak, Armor, Shouting, blood, high quality, absurdres, detailed, perfect face, detailed face, one eye, <lora:Kentaro Miura Style:1>, <lora:Anna_Dittmann_Style_XL:0,7>,
Negative prompt
deformed, blurry, bad quality, worst quality, blurry,
Settings
Sampler
DPM++ 2M Karras
Steps
40
Guidance
7
Seed
3730459309
Size
888x1176
Prompt
(best quality, high quality:1.2),(3d, 3d model, zbrush, high poly),
1boy,solo,bart simpson,
orange t-shirt,blue shorts,blue shoes,
<lora:bart_simpson_sdxl_v1:0.8>,
Negative prompt
worst quality,low quality,2d,illustration drawing,
Settings
Sampler
DPM++ 2M Karras
Steps
10
Guidance
7
Seed
3671375685
Size
1472x2560
Prompt
highly detailed painting, Tim Burton Style, ling1, <lora:Tim_Burton_Painting_Style_SDXL:1.2>, <lora:ling1_2:0.6>, ADDBASE,
a Chinese dragon is attacking from clouds, long wavy dragon, round eyes, detailed face, mountains, ADDROW,
a samurai is attacking, katana above head, high swing, view from behind, village ruins
Negative prompt
nsfw,(worst quality:1.4), (low quality:1.4), (normal quality:1.4), text,logo,watermark,signature, many people, fire, many tails
Settings
Sampler
DPM++ 2M Karras
Steps
30
Guidance
7
Seed
1732105948
Size
1176x888
Prompt
(best quality, high quality:1.3),
ohwx man,3d,3d model,pixar style,high detail,realistic,short beard stubble,bald,mid shot,blue sky,street,white shirt,happy smile,
<lora:homer_simpson_sdxl_v1:1.2>
Negative prompt
(worst quality,low quality:1.2),logo,watermark,signature
Settings
Sampler
DPM++ 2M Karras
Steps
40
Guidance
7
Seed
1178146883
Size
1176x888
Prompt
(masterpiece,best quality, high quality:1.2),highres,photo,realistic,
brian griffin,solo,dog,collar,all fours,looking at viewer,
detailed background,backyard,
<lora:brian_griffin_sdxl_v1:0.6>
Negative prompt
(worst quality,low quality:1.2),lowres,cartoon,2d,leash
Similar Models
AI Models for Image, Video & Text Generation
mohawk18 xl is a base model for the Stable Diffusion XL 1.0 family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Stable Diffusion XL 1.0 models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
More Stable Diffusion XL 1.0 modelsMore Stable Diffusion XL 1.0 Base Models3dsemi-realcomicsbaseartstylesdesigncharacterscartoons
Browse by Model Family
Text / LLM Models
411
Anima Models
509
Chroma Models
14
Flux Models
1,071
Flux 2 / Klein Models
1,237
MiniMax H3 Models
332
Hunyuan Models
198
Ideogram Models
20
Krea2 Models
675
Qwen Models
17
Qwen2 Models
37
Ltx2 Models
151
Stable Diffusion 1.5 Models
6,465
Stable Diffusion XL 1.0 Models
15,432
Zimage Models
804
Wan Models
920

Model
Trends
.ai
© ModelTrends.ai
|
© 2026
|
All Rights Reserved








