import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("LyliaEngine/Pony_Diffusion_V6_XL", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("LyliaEngine/GRIM_v1")
prompt = "(score_9, score_8_up), score_7_up, zPDXL, solo, 1boy, holding, monochrome, weapon, red background, flower, flowly cape, male focus, sword, holding weapon, armor, holding sword, helmet, shoulder armor, gauntlets, spot color, pauldrons, skull, planted, full armor, planted sword, helm,<lora:GRIM_v1:0.9>"
image = pipe(prompt).images[0]GRIM_v1
- Prompt
- (score_9, score_8_up), score_7_up, zPDXL, solo, 1boy, holding, monochrome, weapon, red background, flower, flowly cape, male focus, sword, holding weapon, armor, holding sword, helmet, shoulder armor, gauntlets, spot color, pauldrons, skull, planted, full armor, planted sword, helm,<lora:GRIM_v1:0.9>
- Negative Prompt
- PonyXL_NegScore, negativeXL_D, blurry, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry
- Prompt
- (score_9, score_8_up), score_7_up, zPDXL, solo, looking at viewer, monochrome, greyscale, horns, no humans, black background, monster, multiple horns <lora:GRIM_v1:0.95>
- Negative Prompt
- PonyXL_NegScore, negativeXL_D, blurry, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry
- Prompt
- (score_9, score_8_up), score_7_up, zPDXL, solo, 1boy, meditating, monk, no humans, holding large sword, side view, looking at viewer, looking down, glowing, robot, skull, ribs, skeleton, damaged, bone, feathered wings, black and red theme, limited palette, black and red theme, monochrome, greyscale <lora:GRIM_v1:0.9>
- Negative Prompt
- PonyXL_NegScore, negativeXL_D, blurry, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry
Model description
Did some training on some dark art and metal art. It failed so I merged it with one of my other loras to experiment…and well I thought the results were pretty neat, so I thought I’d share.
be sure to leave a like if it worked for you and you enjoy, helps motivate me to keep publishing my trash 😊
What does this Lora add? Dark and grim stuff, this is a lora where you want to add eldritch abomination into the prompt 😛
Keywords that work good: greyscale, monochrome, limited palette, horror (theme), eldritch abomination, monster, skull, chaotic, silhouette, and such things related.
WARNING: I use a1111 so you may get different results in comfy. (still have to test in comfy.)
Weight : [0.7-1.0]
Source
https://civitai.com/models/470056/grim
Credit
https://civitai.com/user/skumerz
Trigger words
You should use None to trigger the image generation.
Download model
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.
- Downloads last month
- 10
Model tree for LyliaEngine/GRIM_v1
Base model
Bakanayatsu/Pony-Diffusion-V6-XL-for-Anime