<h3 id="heading-39">All Sample image are made Base SDXL_1.0 with Hi-res fix only - (No refiner)</h3>This LoRas enhance/improve image to get more aesthetic/artistic/creative, detailled.<a target="_blank" rel="ugc" href="https://civitai.com/models/124347?modelVersionId=152309">xl_more_art-full_v1</a> this version is little more photorealistic than betas but you can also do anime/illustrations.Weight 0.8-1 is recommended but you ca push more than 1 on less detailled render. ^^This LoRa also contains a training for aerial photos (use "Aerial" keywords), more details here : <a target="_blank" rel="ugc" href="https://civitai.com/articles/2030/how-to-create-realistic-aerial-land-art">https://civitai.com/articles/2030/how-to-create-realistic-aerial-land-art</a><h3 id="heading-16">Please help me :</h3><h3 id="heading-2">Add to your collection and share your creations here !</h3>

The V1 is OUT !!Incredible good !!!!

xl_more_art-full-v1

This is version of xl_more_art-full_v1 with finetuned image of Natural fractalThis model is a little too reactive to certain words like fractal, flower, leaf, tree...It tends to flatten the images.Reactive words :intricate flower wallpaper leaves arafed inspired plant leaf beautiful vibrant skin water flowers pattern veins fractal carline coral spiral macro trees inside palm forest art fractals organic composition droplets lit microscopic symmetrical eye carapic abstract colorful structure fibonacci angle liquid tree covered eyes hyperdetailed symmetry glass bubbles wavy lush explosion energy kolkoutine dome marble grain” broccoli reptile sun ice circle fantasy symmetric river corals atmospheric crown cosmic beeple teal bird staircase deep jan growing space cone metal feathers hurufiyya buds degrees sand jungle ancient tropical wings maze-like winning canvas calatrava grass volumetric waves branches lot salvia cactus neon spines pine lotus iris arbor powerful shell hole cracked floating fern lava fluid frozen bone aster sunlight ocean rosette curvaceous natural honeycomb portal dress circular atmospheric” crystal muted rays post-processing fungus simulacrum bleda petals intricacy various swirl bubble splashes spider hexagonal leather structures cacti extraterrestial brock podedworna farm destiny web ruffles weavers snake effect sunny vnuk drain terracotta hofer spike molecular gloomy brain feather grid jewels horns mathematical shape ivory tesselation coxcomb patterned symmetrically hasselblad oil wood sunrise ink desert worm frill munford curve stalagtites stone sunflower twisting spiralling carnival galactic pineapples angled redwood shore multidimensional scruffy star juice crater overlayed leafy shishk alien oni dragon vine eggs flocks kidney hexagons reef ecosystem horne octan krasovskiy jaegwan sprouts tundra quarter fossilized three spirals cast particles carvings clouds fawcett ribcage heim coin radial gelon curvatures anemones vault line profile furry hastur interlocking caspar blocking seashell painted pretty peppermint motif flora-lush-crater biomechanical cathedral tcolor behance whie brushed candy worms minn velleneuve loosely jewelry meadows adobe gothic carnation shattered quaint snow crypto soap lights krūmiņš mustard abstractions wisconsin hug kavisky lettuce lake mandelbrot cascade sponge hillside hemlocks baer heart slabs botanical pearls twilight seeds dandelion skeleton carapace lavander antennae amazon rainforest glowing resembling stars rust needles kiwi rock

fractal-art.v1

New version (more accurate, more textures)Work good on anime/illustration/character/paint and also PhotorealismAdjusted weight =&gt; 0.7-1.0 are good valueI trained aerial photo and merge in the beta1

xl_more_art-full-beta2

Get images more photorealistic, add detailsThis Lora make images more creative, aesthetic, detailled.<ul><li>weight 0.7 is a good value for cfg scale &gt;= 7</li><li>weight 0.8-1.0 are a good values for cfg scale 5.5-7.0</li><li>weight 0.1-0.3 is a good value for small changes</li><li>Trained with 9 epochs on 6k steps</li></ul>I use realistic images to train this lora

xl_real_beta1

This Lora make images more creative, aesthetic, detailled.<ul><li>weight 0.7 is a good value for cfg scale &gt;= 7</li><li>weight 0.8-1.0 are a good values for cfg scale 5.5-7.0</li><li>weight 0.1-0.3 is a good value for small changes</li></ul>Trained with 8 epochs on 14k stepsArtistic image are used to train this lora

Type	LoRA
Stats	21,081 237,094
Uploaded	Aug 31, 2023
Base Model	SDXL 1.0
Hash	AutoV2 15E31FE2B6

xl_more_art-full / xl_real / Enhancer

All Sample image are made Base SDXL_1.0 with Hi-res fix only - (No refiner)

Please help me :

Add to your collection and share your creations here !