This model is mainly used in concept art design and film and television special effects design, with effects similar to UE5.

Use large models: aZovyaRPGArtistTools_v4.safetensors (this is very important) other models are a bit incompatible

Disadvantages of this model: This lora is the first generation, and the effect is not so perfect. For example, the face may collapse.

So I suggest you must turn on high-resolution restoration. My suggestion here is: R-ESRGAN 4x+ Anime6B

And I used details Lora: more_details.safetensors (this is more important)

This model is marked by Qiuye WD14. You can generate it mainly if your text is smooth. Be sure to type more prompt words, so that the effect will be very good.

Positive prompt words: Pet Legion, blue eyes, no humans, glowing, blurry, animal focus, character design, thick coated art, wearing armor, UHD, 8K, unreal engine rendering, glowing_eye, wearing a cloak, <lora:more_details: 0.9>,artistic work,work of the arts fine,concept art,game artist,art design,necklace,looking at viewer,blurry,realistic,no humans,stand,<lora:cat and dog-000010:0.9>,wearing a mask,gold mask,eyes glow,green eyes,no beard,medium_shot,the whole body,stand,soldier,beautiful and delicate water,the finest grass,very delicate light,nature,painting,water spray,breeze,flowers and grass meadow ,near the water edge,(sunset, starry sky in a circle),randomly distributed clouds,river,splashing water,falling petals,(note that you can imagine the prompt words by yourself)

Reverse prompt words: EasyNegativeV2,ng_deepnegative_v1_75t,(low_quality:1.4),(worst_quality:1.4),(badhandv4:1.1),collage,artist_name,signature,artist_logo,watermark.,FastNegativeV2,EasyNegative,BadNegAnatomyV1-neg,username,worstquality, bad anatomy,out of focus,grayscale,low quality,blurry,bad proportions,monochrome,cropped,normal quality,lowres,watermark,hair,beard,

Number of steps to take: between 25-40 (I recommend 30)

Sampler: Euler a