🚀 ProteusV0.3-Lightning
ProteusV0.3-Lightning採用了字節跳動發佈的全新Lightning方法,能在保證質量和提示理解能力的前提下,實現更快的推理速度。
🚀 快速開始
模型介紹
Proteus是在OpenDalleV1.1基礎上的一次重大升級,它充分利用了OpenDalleV1.1的核心功能,進一步提升了性能。主要改進體現在對提示的響應更加靈敏,以及具備更強的創造力。為了達到這一效果,模型使用了約220,000張帶有GPTV字幕的免版權庫存圖像(其中包含一些動漫圖像)進行微調,並對這些圖像進行了歸一化處理。此外,還通過精心挑選的10,000對高質量AI生成圖像對,運用了直接偏好優化(DPO)技術。
為了實現最佳性能,在訓練過程中會獨立訓練多個低秩自適應(LORA)模型,然後通過動態應用方法有選擇地將它們集成到主模型中。這些技術能夠針對模型的特定部分進行優化,同時避免在學習過程中對其他部分產生干擾。因此,Proteus在描繪複雜的面部特徵和逼真的皮膚紋理方面有顯著提升,同時在各種美學領域,特別是超現實主義、動漫和卡通風格的可視化方面,也保持了出色的表現。
模型設置
使用以下設置可以讓ProteusV0.3-Lightning達到最佳效果:
屬性 |
詳情 |
CFG Scale |
使用1到2的CFG比例 |
Steps |
4到10步可獲得更多細節,8步可獲得更快的結果 |
Sampler |
eular |
Scheduler |
normal |
Resolution |
1280x1280或1024x1024 |
此外,建議在提示詞中使用以下關鍵詞來提升效果:最佳質量、高清、~*~美學~*~
。
如果在構思提示詞時遇到困難,可以使用這個我整理的GPT來幫助優化提示詞:點擊訪問
示例展示
輸入文本 |
輸出圖片 |
Anime Girl holding a sign that says 'Proteus Lighting' |
ComfyUI_08512_.png |
Anime full body portrait of a swordsman holding his weapon in front of him. He is facing the camera with a fierce look on his face. Anime key visual (best quality, HD, ~~aesthetic~~:1.2) |
ComfyUI_08516_.png |
Anime high quality pixel art, a pixel art silhouette of an anime space-themed girl in a space-punk steampunk style, lying in her bed by the window of a spaceship, smoking, with a rustic feel. The image should embody epic portraiture and double exposure, featuring an isolated landscape visible through the window. The colors should primarily be dynamic and action-packed, with a strong use of negative space. The entire artwork should be in pixel art style, emphasizing the characters shape and set against a white background. Silhouette |
ComfyUI_08567_.png |
Super Closeup Portrait, action shot, Profoundly dark whitish meadow, glass flowers, Stains, space grunge style, Jeanne dArc wearing White Olive green used styled Cotton frock, Wielding thin silver sword, Sci-fi vibe, dirty, noisy, Vintage monk style, very detailed, hd |
ComfyUI_08571_.png |
Super cinematic film still of Kodak Motion Picture Film (Sharp Detailed Image) An Oscar winning movie for Best Cinematography a woman in a kimono standing on a subway train in Japan Kodak Motion Picture Film Style, shallow depth of field, vignette, highly detailed, high budget, bokeh, cinemascope, moody, epic, gorgeous, film grain, grainy |
ComfyUI_08578_.png |
in the style of artgerm, comic style,3D model, mythical seascape, negative space, space quixotic dreams, temporal hallucination, psychedelic, mystical, intricate details, very bright neon colors, (vantablack background:1.5), pointillism, pareidolia, melting, symbolism, very high contrast, chiaroscuro |
ComfyUI_08582_.png |
💻 使用示例
基礎用法
import torch
from diffusers import (
StableDiffusionXLPipeline,
EulerAncestralDiscreteScheduler,
AutoencoderKL
)
vae = AutoencoderKL.from_pretrained(
"madebyollin/sdxl-vae-fp16-fix",
torch_dtype=torch.float16
)
pipe = StableDiffusionXLPipeline.from_pretrained(
"dataautogpt3/ProteusV0.3-Lightning",
vae=vae,
torch_dtype=torch.float16
)
pipe.scheduler = EulerAncestralDiscreteScheduler.from_config(pipe.scheduler.config)
pipe.to('cuda')
prompt = "black fluffy gorgeous dangerous cat animal creature, large orange eyes, big fluffy ears, piercing gaze, full moon, dark ambiance, best quality, extremely detailed"
negative_prompt = "nsfw, bad quality, bad anatomy, worst quality, low quality, low resolutions, extra fingers, blur, blurry, ugly, wrongs proportions, watermark, image artifacts, lowres, ugly, jpeg artifacts, deformed, noisy image"
image = pipe(
prompt,
negative_prompt=negative_prompt,
width=1024,
height=1024,
guidance_scale=1,
num_inference_steps=4
).images[0]
📄 許可證
本項目採用GPL-3.0許可證。
支持作者
如果您覺得這個項目有幫助,請通過以下方式支持作者的工作: