Views
No views yet
2.0e-6 for 10 Epochs on 467 images collected from Danbooru. The model is trained using NovelAI Aspect Ratio Bucketing Tool so that it can be trained at non-square resolutions. Like other anime-style Stable Diffusion models, it also supports Danbooru tags to generate images.1from diffusers import StableDiffusionPipeline
2import torch
3
4model_id = "mio/hiten"
5pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
6pipe = pipe.to("cuda")
7
8prompt = "1girl,solo,miku"
9image = pipe(prompt).images[0]
10
11image.save("./miku.png")
(((masterpiece))),(((best quality))),((ultra-detailed)), ((illustration)),floating, ((an extremely delicate and beautiful)),(beautiful detailed eyes),((disheveled hair)),1girl, bangs, black_hair, blue_sailor_collar, blurry, blurry_background, depth_of_field, eyebrows_visible_through_hair, long_hair, looking_at_viewer, parted_lips, sailor_collar, school_uniform, serafuku, shirt, solo, yoroizuka_mizore,medium_chest,colourful_stages,crown,masterpiece,full_body,white_thighhighs,extremely_detailed_CG_unity_8k_wallpaper,solo,1girl,lights
Negative prompt: nsfw,nipples,lowres,bad anatomy,bad hands, text, error, missing fingers,extra digit, fewer digits, cropped, worstquality, low quality, normal quality,jpegartifacts,signature, watermark, username,blurry,bad feet, (((mutilated))),(((((too many fingers))))),((((fused fingers)))),(((extra fingers))),(((mutated hands))),extra limbs,(bad_prompt), (((mutilated))),(((((too many fingers))))),((((fused fingers)))),(((extra fingers)))
Steps: 24, Sampler: DPM2 a Karras, CFG scale: 7, Seed: 3722281017, Size: 512x768, Model hash: 53a39f6a, Model: hiten_epoch10, Batch size: 4, Batch pos: 3, Clip skip: 2, ENSD: 31337