Model
Trends
.ai
Model
Trends
.ai
the best open source ai models
Run Locally
3xpl0r3_dung30n-k.safetensors
218 MB · SafeTensor · bf16
Downloads are served by Civitai. Some creators ask for a Civitai login first.
AD
Install / Run on Graydient
Requires membership
<dungeon-krea2>
About This Model
KREA2 · LoRAs
Created by
arrssenne
on Civitai
by arrssenne - Explore Dungeon Inside the set: - Dragon's hoard cavern - Fey waterfall grotto - Ruined throne room - Flooded prison - Impossible library - Founders' crypt - Dwarven magma forge - Fungal forest cavern - Bridge over an abyss - Serpent temple - Crystal geode cavern - Crossroads synthesis connecting the explored zones
Creator notes
ENGLISH I created 3xpl0r3 dung30n as a photorealistic colossal fantasy dungeon background LoRA built like a continuous exploration. The set moves through 11 radically different locations while keeping one shared art direction: ancient weathered stone, torchlight, low fog, epic scale, and a small carved spiral symbol that quietly binds every room together. Each scene was designed as an empty stage with a wide open center for placing a character, and all 45 training images are free of people and creatures. Inside the set: - Dragon's hoard cavern - Fey waterfall grotto - Ruined throne room - Flooded prison - Impossible library - Founders' crypt - Dwarven magma forge - Fungal forest cavern - Bridge over an abyss - Serpent temple - Crystal geode cavern - Crossroads synthesis connecting the explored zones How to use it: Trigger word: 3xpl0r3 dung30n Recommended weight: 0.8-1.0 Designed for SDXL, Illustrious, Pony, and natural-language generators such as Krea. Example prompt: 3xpl0r3 dung30n, photorealistic, colossal fantasy dungeon, ancient weathered stone, torchlight, low fog, epic scale, crystal geode cavern, refracted purple light beams, wide open center floor space, cinematic lighting, no humans, no creatures Suggested negative prompt: humans, people, body, face, creature, dragon, gore, readable text, logo, watermark, low quality, blurry Usage tips: 1. Use a weight around 0.8 for more freedom from the checkpoint, or move toward 1.0 for a stronger dungeon identity. 2. Describe the desired zone and camera angle explicitly; the training set includes wide, plunging, ground-level, and diagonal compositions. 3. Keep "wide open center floor space" in the prompt when you want a clean area for a character or face-swap workflow. arrssenne FRENCH J'ai cree 3xpl0r3 dung30n comme un LoRA de backgrounds photorealistes representant un donjon fantastique colossal construit comme une exploration continue. Le set traverse 11 lieux radicalement differents tout en conservant une direction artistique commune : pierre ancienne usee, torches, brume basse, echelle epique et un petit symbole en spirale grave qui relie discretement toutes les salles. Chaque scene est concue comme un decor vide avec un large espace central libre pour placer un personnage, et les 45 images d'entrainement ne contiennent ni personne ni creature. Contenu du set : - Caverne du tresor du dragon - Grotte feerique avec cascade - Salle du trone en ruine - Prison inondee - Bibliotheque impossible - Crypte des fondateurs - Forge naine au magma - Caverne de la foret fongique - Pont au-dessus de l'abime - Temple du serpent - Caverne geode de cristal - Synthese du carrefour reliant les zones explorees Comment l'utiliser : Trigger word : 3xpl0r3 dung30n Poids recommande : 0.8-1.0 Concu pour SDXL, Illustrious, Pony et les generateurs en langage naturel comme Krea. Exemple de prompt : 3xpl0r3 dung30n, photorealistic, colossal fantasy dungeon, ancient weathered stone, torchlight, low fog, epic scale, crystal geode cavern, refracted purple light beams, wide open center floor space, cinematic lighting, no humans, no creatures Negative prompt suggere : humans, people, body, face, creature, dragon, gore, readable text, logo, watermark, low quality, blurry Conseils d'utilisation : 1. Utilisez un poids autour de 0.8 pour laisser plus de liberte au checkpoint, ou montez vers 1.0 pour renforcer l'identite du donjon. 2. Decrivez clairement la zone et l'angle de camera desires; le dataset contient des vues larges, plongeantes, au ras du sol et diagonales. 3. Gardez "wide open center floor space" dans le prompt lorsque vous voulez une zone propre pour ajouter un personnage ou utiliser un workflow de face swap. arrssenne
Trigger words
3xpl0r3 dung30n
151 downloads · 19 likes on Civitai
SHA256
08EBB1F88C52BFB423E987945C728B9A2BDDF5E1979BDB32C22E69BAFB7A4732
ModelTrends.ai Model ID
#28018
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
Euler
Steps
12
Guidance
1
Seed
8717178018118
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same flooded forgotten prison, water-level camera between rusted cell bars, stagnant black water near the lens, submerged cells and algae-draped arches, cold teal underwater glow, sparse torch reflections, dripping chains, wet weathered stone, low fog, epic scale, broad open raised stone walkway through the center, small carved spiral on a side cell threshold, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
65592784143915
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, the same colossal dragon's hoard cavern, diagonal view through two massive toppled ancient stone pillars framing the near left and right edges, mountains of gold coins and gemstones banked along the perimeter, giant shed metallic dragon scales scattered as traces only, smoke wisps rising to a cracked cavern ceiling, braziers, ancient weathered stone, torchlight, low fog, epic scale, rich golden magical glow, cinematic lighting, strong diagonal three-quarter composition, wide open empty center floor space, layered depth toward a distant side arch, one small discreet spiral symbol carved into a toppled pillar, adventurer's discovery atmosphere, realistic depth of field, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
153447177640896
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same impossible library, plunging top-down view into the central vertical well from a high balcony, concentric shelves and unsupported spiral staircases descending through immense depth, chained closed grimoires, floating candles, weathered stone, warm white-gold glow, dust fog, open empty circular floor visible far below, small carved spiral on the balcony parapet, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
213746147477101
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same dwarven magma forge, plunging view from an overhead crane over lava channels, giant side anvils and frozen bronze gears, blackened weathered stone, heat shimmer and smoke, lava-orange magical glow, epic scale, broad open empty stone work platform at the center, small carved spiral on the crane housing, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
379844792457339
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, the same colossal ruined throne room, plunging view from the broken edge of a high collapsed gallery, looking steeply downward across shattered columns toward the distant empty throne, broken crown visible as a small object, cracked mosaic floor, oblique daylight beams through roof breaches, torchlight, drifting dust, low fog, ancient weathered stone, epic vertical scale, subtle cool-white magical glow in mosaic fissures, cinematic lighting, broad open empty mosaic floor visible in the central lower hall, strong height and asymmetry, small discreet spiral symbol carved into the broken gallery parapet, adventurer's discovery atmosphere, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
388177902266247
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same fungal forest cavern, diagonal view beneath luminous violet mushroom gills, root columns receding obliquely, floating spores, violet haze, weathered stone, low fog, epic scale, broad open empty stone clearing at center, small carved spiral on a side root pillar, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
152117792439023
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same impossible library, level camera on a suspended bridge made of massive bookshelf sections crossing the vertical shaft, chained closed grimoires, floating candles, unsupported staircases in the distance, weathered stone, warm white-gold glow, dust fog, epic scale, broad open empty stone landing in the center of the bridge, small carved spiral on a side rail block, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
Settings
Sampler
Euler
Steps
12
Guidance
1
Seed
136432014348155
Size
896x1152
Prompt
3xpl0r3 dung30n, photorealistic, same abyss bridge, low-angle diagonal view from the bridge edge toward the far monumental gate, parapet and hanging vines framing the sides, cold blue abyss glow below, weathered stone, torchlight, low fog, epic scale, broad open empty bridge center, small carved spiral on a low edge block, cinematic lighting, no humans, no silhouettes, no bodies, no creatures, no gore, no readable text, no logos, no watermarks
Negative prompt
worst quality, low quality, text, deformed, bad anatomy, bad hand, blurry, (watermark), artist signature, artist name
AI Models for Image, Video & Text Generation
dungeon krea2 is a LoRA for the Krea2 family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Krea2 models, check its heat score to see how it is trending, and open it on PirateDiffusion or BitVector to try it.
Browse by Model Family
Browse all models