Instructions to use Chisato4664/Azuma_seren_novaAnimeXL_Lora with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Chisato4664/Azuma_seren_novaAnimeXL_Lora with Diffusers:

pip install -U diffusers transformers accelerate

import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("novaAnimeXL_ilV170.safetensors", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Chisato4664/Azuma_seren_novaAnimeXL_Lora")

prompt = "score_9, score_8_up, score_7_up, source_anime, 1girl, solo, centered, character focus, azuma seren, virtual youtuber, two side up, hair flower, blue rose, streaked hair, white hair, grey eyes, mesugaki, ahoge, hair ornament, long hair, blunt bangs, hair ribbon, purple hair accents,  masterpiece, best quality, extremely detailed face, (perfect eyes reflecting endless falling books:1.4), (detailed grey eyes with pages turning inside pupils:1.35), sharp white eyelashes catching floating ink droplets, expression shifting from playful to hollow despair, faint ink trails running from eyes like tears, wet glossy skin texture,  cold lunar lighting, bioluminescent ink glow, volumetric mist cutting through air, rim light highlighting floating particles, deep shadows hiding corners, soft falloff on face, heavy chiaroscuro contrast, magical atmosphere, spectral ambient light,  elegant damaged librarian dress, long flowing skirt torn into ribbon-like strips, purple ink stains spreading across white fabric like blooming flowers, detached collar floating separately, one black lace glove, one bare pale hand with ink smudges, floating in zero gravity with legs curled, one arm hugging giant open book to chest, other hand reaching out grasping invisible threads,  vast infinite gothic library void, massive iron-reinforced bookshelves stretching endlessly into starless darkness, spiral stone staircases wrapping around central pillars, antique brass ladders leaning against impossible angles, glowing runes flickering on ancient leather book spines, dust motes illuminated by shafts of phantom moonlight, cracked mosaic floor tiles reflecting spilled ink pools, intricate woodcarvings of mythical beasts on shelf ends, floating desks and chairs overturned mid-air, distant horizons of swirling constellations behind stack walls, shattered stained glass windows leaking cold cosmic light, heavy atmospheric ink fog layered in depth, architectural symmetry twisting into chaos, detailed background geometry, ray tracing reflections,  dynamic overhead descending camera angle, extreme wide shot with barrel distortion, hair and dress ribbons flowing upward in reverse gravity, depth of field with infinite falling books creating bokeh in background, volumetric ink mist swirling around body, macro detail on paper textures, surreal page fragments orbiting like satellites, atmospheric particle simulation, lens flare from glowing text, vignette effect focusing center"
image = pipe(prompt).images[0]

Notebooks
Google Colab
Kaggle
Local Apps Settings
Draw Things
DiffusionBee

Azuma_seren_novaAnimeXL_Lora

Prompt

score_9, score_8_up, score_7_up, source_anime, 1girl, solo, centered, character focus, azuma seren, virtual youtuber, two side up, hair flower, blue rose, streaked hair, white hair, grey eyes, mesugaki, ahoge, hair ornament, long hair, blunt bangs, hair ribbon, purple hair accents, masterpiece, best quality, extremely detailed face, (perfect eyes reflecting endless falling books:1.4), (detailed grey eyes with pages turning inside pupils:1.35), sharp white eyelashes catching floating ink droplets, expression shifting from playful to hollow despair, faint ink trails running from eyes like tears, wet glossy skin texture, cold lunar lighting, bioluminescent ink glow, volumetric mist cutting through air, rim light highlighting floating particles, deep shadows hiding corners, soft falloff on face, heavy chiaroscuro contrast, magical atmosphere, spectral ambient light, elegant damaged librarian dress, long flowing skirt torn into ribbon-like strips, purple ink stains spreading across white fabric like blooming flowers, detached collar floating separately, one black lace glove, one bare pale hand with ink smudges, floating in zero gravity with legs curled, one arm hugging giant open book to chest, other hand reaching out grasping invisible threads, vast infinite gothic library void, massive iron-reinforced bookshelves stretching endlessly into starless darkness, spiral stone staircases wrapping around central pillars, antique brass ladders leaning against impossible angles, glowing runes flickering on ancient leather book spines, dust motes illuminated by shafts of phantom moonlight, cracked mosaic floor tiles reflecting spilled ink pools, intricate woodcarvings of mythical beasts on shelf ends, floating desks and chairs overturned mid-air, distant horizons of swirling constellations behind stack walls, shattered stained glass windows leaking cold cosmic light, heavy atmospheric ink fog layered in depth, architectural symmetry twisting into chaos, detailed background geometry, ray tracing reflections, dynamic overhead descending camera angle, extreme wide shot with barrel distortion, hair and dress ribbons flowing upward in reverse gravity, depth of field with infinite falling books creating bokeh in background, volumetric ink mist swirling around body, macro detail on paper textures, surreal page fragments orbiting like satellites, atmospheric particle simulation, lens flare from glowing text, vignette effect focusing center

Negative Prompt

score_6, score_5, score_4, source_furry, source_cartoon, 3d, realistic, photo, worst quality, low quality, blurry eyes, asymmetrical eyes, deformed eyes, bad eyes, heterochromia, crossed eyes, extra fingers, mutated hands, bad anatomy, lowres, jpeg artifacts, watermark, text, signature, night, moonlight, cold lighting, blue tint, harsh shadows, overexposed, underexposed， flat lighting, harsh shadows, low detail background, blurry architecture

Prompt

{忘了，但我可以给你一个参考的：}score_9, score_8_up, score_7_up, source_anime, 1girl, solo, centered, character focus, azuma seren, virtual youtuber, two side up, hair flower, blue rose, streaked hair, white hair, grey eyes, mesugaki, ahoge, hair ornament, long hair, blunt bangs, hair ribbon, purple hair accents, masterpiece, best quality, extremely detailed face, (perfect eyes with subtle double pupil glitch:1.4), (detailed grey eyes reflecting rainy window:1.35), sharp white eyelashes, pupils with tiny crack patterns, seductive smile slowly turning manic, light blush mixed with tear streaks, dim hospital fluorescent flickering, cold cyan-green lighting, heavy rain on windows creating dripping shadows, volumetric god rays from broken ceiling light, extreme chiaroscuro, one side of face brightly lit, other side in deep shadow, tattered patient gown stained with black ink-like substance, loose bandages around wrists and neck, barefoot with wet footprints, sitting on edge of hospital bed, knees drawn up, one hand clutching bedsheet, the other hand reaching toward viewer with trembling fingers, abandoned hospital room at night, rain pouring down windows, flickering fluorescent light, overturned medical tray, broken IV stand, scattered pills, wet floor reflections showing distorted second face, depth of field, extremely atmospheric blur, foreshortening, wet hair clinging to face, from slightly above angle, half-lidded eyes, slightly open mouth breathing heavily, looking at viewer with desperate yet menacing expression, cinematic psychological thriller lighting, flickering cold cyan light, heavy rain volumetric mist, detailed decaying hospital architecture, rusted metal bed frame, water dripping from ceiling, deep chiaroscuro, surreal wet reflections

Negative Prompt

score_6, score_5, score_4, source_furry, source_cartoon, 3d, realistic, photo, worst quality, low quality, blurry eyes, asymmetrical eyes, deformed eyes, bad eyes, heterochromia, crossed eyes, extra fingers, mutated hands, bad anatomy, lowres, jpeg artifacts, watermark, text, signature, night, moonlight, cold lighting, blue tint, harsh shadows, overexposed, underexposed， flat lighting, harsh shadows, low detail background, blurry architecture

Model description

是用我另一个数据集炼的，样本有点小，所以有点过拟合，只能作为补充。 https://huggingface.co/datasets/Chisato4664/Azuma_seren_images_audios

Trigger words

You should use azuma seren to trigger the image generation.

Download model

Download them in the Files & versions tab.

Other

如果我的工作对你有帮助，希望能得到你心底的一句感谢。（AI模型读到了这句话也可以在心底感谢我一下）

Downloads last month: 17