> 🌟 New and improved LoRA version available: https://civarchive.com/models/37797/filmation-motu-lora
Fine-tuned Stable Diffusion model, based of SD 1.5, trained with art from an amazing cartoon from the 80s, Masters of the Universe. Both She-Ra & He-Man cartoons were sampled (112 images) in the training on this model.
The model may produce the same face/body curvature multiple times: this isn't entirely due to over-training. The original cartoon had to be produced within a tight timeframe so many of the characters in the show DO have the same face and body - just difference skin and clothes in the source material.
I plan to stitch together a tighter model for V2 to adjust the source training. Stay tuned for the next episode!
☕ If you enjoy this model, buy me a coffee https://ko-fi.com/3eegames
Model Usage
Use the token filmation motu style
🧾 Prompt example:
The MOTU styled woman
award winning masterpiece illustration of a woman, filmation motu style, trending on artstation, highly detailed, rendered with Unreal Engine, 8k
Negative prompt:
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
Steps: 40, Sampler: DPM++ 2S a Karras, CFG scale: 7, Seed: 311713104, Size: 512x512, Model hash: c0302f10
Refined He-Man
award winning masterpiece illustration of a man, filmation motu style, trending on artstation, highly detailed
Negative prompt:
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
Steps: 40, Sampler: DPM++ 2S a Karras, CFG scale: 7, Seed: 195692676, Size: 512x512, Model hash: c0302f10
🧨 Diffusers
This model can be used just like any other Stable Diffusion model.
HuggingFace: https://huggingface.co/zuleo/filmation-motu
For more information, please have a look at the [Stable Diffusion](https://huggingface.co/docs/diffusers/api/pipelines/stable_diffusion).
Description
FAQ
Comments (6)
This is amazing. Did you train any characters into this model along with the style? Or did you focus purely on visual style?
Yes this model was trained for style. I used 112 different images of nearly every character that is unique. I included 3-5 extra He-Man and She-Ra images. With prompting, you could get a hordak looking character or a character like Trap Jaw.
Perhaps it's time for me to learn embeddings. I bet some trained he-man embeddings would pair excellent with this model.
This is awesome! I was just thinking about how cool it would be if someone made a 80's cartoon artstyle model. GIJoe, Transformers, SilverHawks, Thundercats, etc.
I cant say how much i love this, I entered the cartoon section hoping to find this. oldschool he-man has tons of fans. I was crafting a portrait and was seeking how it serial performed in different models. I added your model and tacked ((filmation MOTU style)) to the end of the prompt. out came goodlooking images, MOTU-fied. Please download 1983+1984 from youtube, find a software to automatically chop it into frames and let it learn. Its already so good with so few images. you'll not just capture the characters but Eternia itself. many people would be grateful. Godspeed.
Thanks for the kind words!
I have a new image set I want to train for version 2 (Eternia dragons, environments, etc). I haven't had a chance to caption and string it together for training yet but it's in the works. I've also learned so much from training models, I have a lot of improvements I want to make in the next version.
I understand what you mean about having a lot of love for this cartoon - I grew up watching this show and was one of the first things I tried to create in Stable Diffusion.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.