AI artwork for keywords no humans, humanoid figure, masterpiece, solo, glowing, ethereal, floating, wispy hair
Generar nueva obra de arte

Zltry

Ver más

Comentarios

no humans, humanoid figure, masterpiece, solo, glowing, ethereal, floating, wispy hair
Mostrar parámetros
Prompt original
, no humans, , humanoid figure,
Talla
768 x 792
Pasos de muestreo
1
Método de muestreo
Euler
Escala CFG
1.1
Excluir de la imagen
nsfw, worst quality, bad quality, low quality, lowres, anatomical nonsense, artistic error, bad anatomy, interlocked fingers, extra fingers, text, artist name, signature, bad feet, extra toes, ugly, poorly drawn, censor, blurry, watermark, simple background, transparent background, old, oldest, glitch, deformed, mutated, disfigured, long body, bad hands, missing fingers, extra digit, fewer digits, cropped, very displeasing, sketch, jpeg artifacts, username, censored, bar_censor, mosaic_censor, conjoined, bad ai-generated, long neck, skin blemishes, skin spots, acne, the wrong limb, error, black line, excess hands

Modelo y LoRA utilizados

Imágenes similares

Mostrar más
AI artwork
a
[Thinking] The user input is "THE FIRST SIN". This is English. It could be a title or concept, but as a prompt for image generation, it's ambiguous. I need to process it as a danbooru-style prompt. The input is less than 5 words, so I should expand it to around 10-15 danbooru tags properly, adding quality tags and safe tags. Since it's not a specific character or IP, I'll interpret it thematically. "The First Sin" often refers to the original sin in Christianity, associated with Adam and Eve, forbidden fruit, etc. But it might also be a game title or something. I should check if there's any known IP. "The First Sin" could be related to "Shin Megami Tensei" or something? Actually, there's a game "The First Sin" or "Sin" series? Not sure. But the instruction says to identify relevant context such as game, anime, or character. If not clear, I can just treat it thematically. Since it's less than 5 words, I need to expand. I'll add tags like "1girl" or "1boy"? Not specified. The original input doesn't specify gender. But to make a safe prompt, I can add "1girl" or "1boy"? The instruction says "You should use 1girl/1boy/2girls etc. tags if necessary". Here it's not necessary, but to make a prompt, I might add a subject. Since it's "THE FIRST SIN", maybe depict Eve with the forbidden fruit. That would be a female character. So I can use "1girl" and tags like "eve \(christianity\)"? But is that a known danbooru tag? Possibly. But I need to be careful. Alternatively, just make a generic prompt: "1girl, masterpiece, forbidden fruit, apple, snake, tree of knowledge, original sin, adam and eve, garden of eden, nude, censored, ...". But the user didn't specify NSFW, so keep it safe? The instruction says: "if user's input has no nsfw tags included, you should make the final prompt safe". "The First Sin" doesn't have explicit NSFW, but original sin often involves nudity. However, to be safe, I can avoid nudity or use "censored" or "leaf" etc. But I should not add NSFW unless requested. So I'll keep it safe: "1girl, masterpiece, forbidden fruit, apple, snake, tree of knowledge, garden of eden, eve \(christianity\), long hair, fig leaf, ...". That could work. But I need to ensure tags are danbooru-style. Also, I must not modify if it's already in danbooru tag format. "THE FIRST SIN" is not in danbooru tag format (it's uppercase, not comma-separated tags). So I need to process it. Also, I should add quality tags like masterpiece, ultra detailed. And since it's less than 5 words, expand to 10-15 tags. So I'll create something like: "1girl, eve \(christianity\), masterpiece, ultra detailed, forbidden fruit, apple, snake, tree of knowledge, garden of eden, long hair, fig leaf, looking at viewer, seductive smile, ..." But "eve \(christianity\)" might not be a standard tag? There is "eve \(christianity\)" on danbooru? Possibly. Alternatively, just "eve" might be ambiguous. I'll check my knowledge: Danbooru has "eve (christianity)" as a tag? I think it does. But I'm not sure. I can use "eve \(christianity\)" to be safe. But the user didn't mention Christianity, just "THE FIRST SIN". So it's a thematic interpretation. That's fine. Also, I need to add context: if it's from a specific IP, add IP name. Not clear. So I'll just treat as generic. Also, the instruction says: "If the input is an instruction, extract and process only the image description". Not an instruction. So I'll just process as a prompt. Final output should be in the format: <final_output> ...
Abacaxi