a badass combo of Alien and Predator, with a pinch of mechanical creatures and various body parts. Base model (Alien + Mech) trained in One trainer with EMA, LOKR Predator in Kohyass.

ALIEN PREDATOR MECH COMBO

alienPREDATORMECH_v10.yaml

alienPREDATORMECH_v10.safetensors

v1.0

Predator Lokr

The dataset for this model was created quite by accident. I trained the embedding with images from the Dead Space game, and the images generated during training deviated a lot from the original images, but they are so good that I couldn't resist training the model using them. Training was done in One Trainer with EMA enabled, which is the only thing really missing in Kohyass.

SciFi Surreal

scifiSurreal_v10.yaml

scifiSurreal_v10.safetensors

Here's json metadata file with all captions used for training. I am truly awesome, i know <a target="_blank" rel="ugc" href="https://drive.google.com/file/d/14D0ulhEtfg3G2XhftMxoXE-mTLjN233E/view?usp=sharing">https://drive.google.com/file/d/14D0ulhEtfg3G2XhftMxoXE-mTLjN233E/view?usp=sharing</a>...now, would you kindly post some crazy pics, please.EDIT2: links to embeddings and hypernetwork <a target="_blank" rel="ugc" href="https://drive.google.com/drive/folders/11XRrsqaLnQdXoo59GdqB353hWYjEtjb7?usp=sharing">https://drive.google.com/drive/folders/11XRrsqaLnQdXoo59GdqB353hWYjEtjb7?usp=sharing</a> <a target="_blank" rel="ugc" href="https://drive.google.com/drive/folders/1csY1ig4hgAnzrdgSXUm5YnL9EzqhGwhB?usp=sharing">https://drive.google.com/drive/folders/1csY1ig4hgAnzrdgSXUm5YnL9EzqhGwhB?usp=sharing</a> Hypernetwork is trained in Monkey patch extension, so you need to have that installed in order to use this HyperN************************************************************************************** <a target="_blank" rel="ugc" href="http://zwx-2000.pt">zwx-2000.pt</a> is embedding (positive), scrnstl_<a target="_blank" rel="ugc" href="http://v4.pt">v4.pt</a> in Hypernetwork (should also be placed in positive prompt), and <a target="_blank" rel="ugc" href="http://negativefixer.pt">negativefixer.pt</a> is negative embedding. If you mix all of that into one prompt, you get a tasty stew made of skin, fingernails, exposed flesh and lost hopes.

SCORN

scorn_v10.yaml

scorn_v10.safetensors

Model trained on approximately 1000 carefully selected images (people, monsters, objects, buildings, animals and plants). I've used the One trainer, which I really like.Because i'm awes../. I'm giving you all the captions... <a target="_blank" rel="ugc" href="https://drive.google.com/file/d/1Y9b5RmbgHBt3CsyeF7ZD682nxWKbrDfT/view?usp=sharing">https://drive.google.com/file/d/1Y9b5RmbgHBt3CsyeF7ZD682nxWKbrDfT/view?usp=sharing</a>

Dark Pureness

darkPureness_v10.yaml

darkPureness_v10.safetensors

V2 notes; Added 10k more images to the base and trained for 100 epochs. Keywords to use are GTv2 and goldentrig. You'll see the difference in the images i posted for the model for a few comparisons. you'll notice i just added the keyword at the beginning of the prompt and boom; better quality :)Training: 250 hours on a RTX 3060. Used OneTrainer DAdapt Lion this time :). Recommended params:Here's my first SD1.5 model. GoldenTrig. The inspiration for this model comes from a prompting strategy I developed for SDXL and my mix HouseOfFreyja. You'll notice most of my prompts include the following tags "(quality{[intricate_insane_exponential_details_8k_resolution_volumetric_lighting_depth_of_field_FullFrame_36x24mm_Nikon_D850_sensor_grain_texture]}),(composition{[.707_trigonometric_discretization_scalar*1.618033988749895→lighting_angle_quantization_color_reflection_modeling_aesthetic_spacing_principles*scene_structure]}),". So what I decided to do was find a lot of prompts. I mean a lot. <a target="_blank" rel="ugc" href="https://huggingface.co/datasets/daspartho/stable-diffusion-prompts">https://huggingface.co/datasets/daspartho/stable-diffusion-prompts</a>. I unpacked the parquet file and got 3.2gb of text worth of prompts. 1.8 millionish. Then I appended those tags to every single one of those prompts. This base mix of this model is (INSERT MIX LIST HERE LATER DUE TO ADHD). This model is the first 10k images generated from the prompts. I used <a target="_blank" rel="ugc" href="https://github.com/Nerogar/OneTrainer">https://github.com/Nerogar/OneTrainer</a> to train my model using Prodigy and Rex for the optimizer and LR scheduler. Additionally used xformers.Recommend params:Schedulers: Euler a, DPM++SDE, DPM++3M SDE, RESTART,Steps:30-50 (66 for RESTART)CFG:5-9 (higher CFG=more steps)Happy generating!!

Files with SHA256: 299b69c5769a6e48a0cb7b56051bb8199f64e2c35e72e89ea8ff253aadcc9cbf