What can I create with this?
Hobbit house-inspired interiors, with large windows, nice views, luxury furniture, and dark wood and leather material.
Quick Example
Positive prompt
(Professional 3D rendering:1.3) of (Ultrarealistic:1.3) magicalinterior style, bathroom, bathing room, plant, scenery, window, tree, nature, bathtub, forest, mirror, bath, faucet, water, house interior, indoors, cottage interior, treehouse interior, dark, gothic, Goth, luxury, fantasy, magic, Harry Potter, Slytherin, perfect composition, award-winning,CGSociety,ArtStationNegative prompt
ng_deepnegative_v1_75t, bad-artist, (worst quality, low quality:1.4), multiple views, multiple panels, blurry, watermark, letterbox, text, character, human, people, 1girl, 1boy,((Wireframe)),Polygons,Screenshot,Character design,Software,UI,watermark,signatureExample Settings
768x512 px canvas size, CFG scale 7 and 70 steps on Deliberate.
Complete workflow
Prompt structure
[your medium], magicalinterior style, [Subject], [Descriptors]
Medium
You can use "professional 3d rendering" or set Image Type to Random in StylePile (highly recommended extension).
Subject
Use one of these subjects or copy-paste them to the [X] field of Style Pile and use [X] in your prompt formula instead of [Subject]. You can also generate your original subjects in ChatGTP, using the workflow described here.
kitchen, kitchen decor, kitchen design, window, scenery, plant, table, bottle, sunlight, chair
bathroom, bathing room, plant, scenery, window, tree, nature, bathtub, forest, mirror, bath, faucet, water
living room, livingroom, plant, candle, window, scenery, couch, table, candlestand, cup, potted plant, teapot, sleeping, chair, stained glass
potions lair, candle, city, scenery, window, fantasy, witchcraft, alchemy, alchemist lab, candlestand, bottle, chandelier, table, shelf
library, no scenery, books, bookshelf, window, plant, tree, lamp, table, candle, flower, chair
Descriptors
The last part of the positive prompt is descriptive tags. Use one of these or make your own:
house interior, indoors, cottage interior, treehouse interior, dark, gothic, Goth, luxury, fantasy, magic, Harry Potter, Slytherin, Lord of the rings, LOTR
house interior, indoors, cottage interior, treehouse interior, light, bright, luxury, fantasy, magic, Harry Potter, Slytherin, perfect composition, award-winning
house interior, indoors, cottage interior, treehouse interior, dark, gothic, Goth, luxury, fantasy, magic, Harry Potter, Slytherin, perfect composition, award-winning, Dirk Crabeth, unreal engine 5 highly rendered, a detailed matte painting, gothic art
Negative prompt
Use two of your favorite negative embeddings (in my case Deep Negative and Bad Artist) plus negative words from the model you are using. My case:
ng_deepnegative_v1_75t, bad-artist, (worst quality, low quality:1.4), multiple views, multiple panels, blurry, watermark, letterbox, text, character, human, people, 1girl, 1boy
Recommended models
Good results on Deliberate, The Allys Mix III, and Surreal Steampunk AI (without trigger words). Style fails to deliver with Anime models, such as Grapefruit.
Upscaling
Examples are not upscaled but you can do it with a denoising of 0.2 and your favorite upscaler in IMG2IMG.
Description
Trained with 55 high-quality Midjourney images with variable speed (0.05:100, 0.02:200, 0.01:600, 0.005:2000, 0.002:5000, 0.001:10000, 0.0005) up to 40 000 steps.
FAQ
Comments (5)
If I dont add other color prompt to it,all output is green style..
then add other color prompts
used till a week ago without problems, now this LORA give error and never activate itself. Due to some automatic1111 updates? Very strange but is the only LORA to do so.
The error is:
activating extra network lora with arguments [<modules.extra_networks.ExtraNetworkParams object at 0x7f99f4ebb970>]: ValueError Traceback (most recent call last): File "/content/stable-diffusion-webui/modules/extra_networks.py", line 75, in activate extra_network.activate(p, extra_network_args) File "/content/stable-diffusion-webui/extensions-builtin/Lora/extra_networks_lora.py", line 23, in activate lora.load_loras(names, multipliers) File "/content/stable-diffusion-webui/extensions-builtin/Lora/lora.py", line 170, in load_loras lora = load_lora(name, lora_on_disk.filename) File "/content/stable-diffusion-webui/extensions-builtin/Lora/lora.py", line 108, in load_lora key, lora_key = fullkey.split(".", 1) ValueError: not enough values to unpack (expected 2, got 1)
This is not a Lora, this is textual inversion. Put it in the embeddings folder and prompt the trigger word.
@Peaksel ok, resolved! Thanks. I was a bit confused...
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.