%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
ZH
Natalie Portman Sdxl Flux
模型简介
该模型是一个定制化的LoRa训练模型，专注于生成高质量的女性肖像图像，支持从照片级真实感到动漫风格的多种艺术表现。模型特别擅长处理性感、名人风格的肖像，并能适应不同的背景和场景需求。
模型特点
多风格支持
能够生成从照片级真实感到动漫风格、油画风格等多种艺术表现形式的图像。
高质量肖像生成
特别擅长生成高质量的女性肖像，注重细节如妆容、饰品和表情。
场景适应性强
能够根据提示词生成各种背景场景，从夜店到自然风光都能很好表现。
名人风格处理
能够处理特定名人（如娜塔莉·波特曼）的风格特征，生成具有辨识度的肖像。
模型能力
文本到图像生成
风格转换
肖像生成
场景生成
艺术风格模仿
使用案例
艺术创作
肖像艺术
为艺术家提供创作灵感或直接生成可用于艺术项目的肖像作品。
高质量、风格多样的肖像图像
概念设计
为游戏、电影等娱乐产业生成角色概念图。
风格统一、细节丰富的角色设计
商业应用
广告素材
快速生成符合品牌调性的广告人物形象。
专业级的人物摄影效果
社交媒体内容
为网红、博主生成独特的个人形象照片。
Instagram风格的高质量自拍
娱乐
粉丝艺术
生成名人或虚构角色的粉丝艺术作品。
具有辨识度的名人风格肖像
角色扮演素材
为COSPLAY爱好者提供角色参考图像。
详细的服装和妆容参考
🚀 娜塔莉·波特曼（SDXL + FLUX）
本项目基于SDXL和FLUX技术，利用娜塔莉·波特曼的图像数据训练了多个版本的LoRA模型，可用于生成不同风格和场景下的娜塔莉·波特曼图像，为文本到图像的生成提供了多样化的选择。
🚀 快速开始
模型下载
此模型的权重以Safetensors格式提供。你可以在Files & versions标签页中下载。
使用🧨 diffusers库调用模型
from diffusers import AutoPipelineForText2Image
import torch
device = "cuda" if torch.cuda.is_available() else "cpu"
pipeline = AutoPipelineForText2Image.from_pretrained('black-forest-labs/FLUX.1-dev', torch_dtype=torch.bfloat16).to(device)
pipeline.load_lora_weights('Keltezaa/natalie-portman-sdxl-flux', weight_name='natportman_2004_local_164_v2-000005.safetensors')
image = pipeline('High quality passport photo of a woman wearing a suit and tie looking directly at the camera with her mouth closed and a neutral expression. She is also wearing a delicate gold chain and some understated diamond earrings. ').images[0]
更多详细信息，包括LoRA的加权、合并和融合，请查看diffusers中加载LoRA的文档。
✨ 主要特性
多版本模型
- FLUX v.2：基于FLUX.1 [dev]，使用84张带有GPT - 4字幕的娜塔莉·波特曼照片进行训练。在FLUX 1.dev（完整）、FLUX fp8和FLUX nf4上进行了测试。建议使用强度为0.8 - 1.2，蒸馏CFG 3.5和CFG 1.0（无负提示），Clipskip 1。
- 正向提示示例：
{Artstyle, Character and scene description in usual FLUX fashion}, <lora:natportman_2004_local_164_v2-000005:1.0>
- 正向提示示例：
- FLUX v.1 (Portman in 2004)：基于FLUX.1 [dev]，使用100张2004年带有GPT - 4字幕的娜塔莉·波特曼照片进行训练。在FLUX 1.dev（完整）、FLUX fp8和FLUX nf4上进行了测试。建议使用强度为0.8 - 1.2，蒸馏CFG 3.5和CFG 1.0（无负提示），Clipskip 1。
- 正向提示示例：
{Artstyle, Character and scene description in usual FLUX fashion}, <lora:Natalie_Portman_2004_FLUX_epoch_16:1.1>
- 正向提示示例：
- v6.0 2004：基于Juggernaut X，使用230张2003 - 2004年的娜塔莉·波特曼照片进行训练。在Juggernaut X、Juggernaut v7、RealismEngine 2、RealVisXL3和AlbedoBase 2.0上进行了测试。使用关键词
"ntxprtman"，建议使用强度为1.0 - 1.1，CFG 5.0 - 7.0，Clipskip 1，步数为10 - 40。- 正向提示示例：
{Artstyle}, {Character and scene description}, elxolsn, <lora:natportman_2003_juggerX_xl_1_wocap-natxprtmn-000149:1.0> - 负向提示示例：
ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child
- 正向提示示例：
- v5.0 Juggernaut X：为Juggernaut X训练，使用270张娜塔莉·波特曼的照片。建议与Juggernaut X、Juggernaut v7或RealismEngine 2一起使用，与Juggernaut X配合效果最佳。使用关键词
"natxportman"。- 正向提示示例：
{Artstyle}, {Character and scene description}, natxportman, <lora:natportman_gpt4_juggernautX_2_wocap-merger_21_65_83_04_02_04-natxportman:1.0> - 负向提示示例：
ugly, deformed, airbrushed, photoshop, rendered, (multiple people), child
- 正向提示示例：
- v4.0 80mb：基于137张娜塔莉·波特曼的照片训练，改进的LoRA模型在保持相似度的同时大幅减小了尺寸。在提示开头使用关键词
"natxportman"，建议LoRA使用强度约为1.1。 - v3.1：基于v3.0的200张图像数据集中的一个子集进行训练，该子集具有更好的平衡性和字幕（更精细的GPT - 4视觉提示）。LoRA强度为0.8 - 1.2。
- v3.0：基于200张图像重新训练，采用改进的LoRA训练方法和模型，在保持相似度和GPT - 4视觉字幕的同时提供更高的灵活性。权重保持在1.05 - 1.1。
- v2.0：基于SDXL 1.0基础，使用270张娜塔莉·波特曼的照片训练LoRA。该LoRA也在Juggernaut XL 3.0上进行了测试。除了“...2004...”生成的图像外，大多数示例图像是使用Juggernaut创建的。
- v1.0：基于SDXL 1.0基础，使用45张娜塔莉·波特曼的照片训练LoRA。ComfyUi的推荐强度为1.1。
📄 许可证
本模型使用定制的LoRA训练许可证。
模型示例展示
以下是一些提示文本及其生成的图像示例：
| 提示文本 | 生成图像链接 |
|---|---|
| High res waist up portrait photo of a woman directly facing the camera with eye - makeup featuring glittery eye - shadow and clear glossy lip - gloss. She is looking at the camera with a slight seductive look. She is wearing a thin string - like black choker and hoop earrings. In the background is a nightclub scene out of focus. | 33238788.jpeg |
| Use modern vector graphics flat design art with vibrant colors with a limited color pallete and simple shapes to create an illustration of a woman wearing flowers in her hair looking at the viewer. | 33198302.jpeg |
| High res waist up portrait photo of a woman directly facing the camera with eye - makeup featuring glittery eye - shadow and clear glossy lip - gloss. She is looking at the camera with a slight seductive look. She is wearing a thin string - like black choker and hoop earrings. In the background is a nightclub scene out of focus. | 33237679.jpeg |
| Contact sheet with 4 images of a woman with a messy short bob cut wearing a black turtleneck. | 33198319.jpeg |
| Detailed realistic drawing of the face of a woman facing the viewer but the left half of her face is skull while the right side of her face is normal. She has no hair on the left side of her face only a skull and also her eye - socket is hollow on the left side. | 33198332.jpeg |
| High quality studio photo of a sexy woman with beautiful eyes and her hair in a ponytail wearing a tight shirt and jeans with her midsection visible and a thin tight choker with a metal ring attached to the front of it around her neck looking at the viewer. On the plane white wall behind her is a blue analog clock which shows the time to be 2:30. Bellybutton is visible. | 33198350.jpeg |
| ((Retro 90s Anime style illustration)) drawing of a closeup on face of woman wearing a long black trench - coat futuristic city in the background in the style of the 90s anime Akira. | 33198360.jpeg |
| Instagram photo of a woman with freckles on her face in the winter outdoors silhouette illumination of her hair. Taken with a ProPhoto iPhone camera. | 33198368.jpeg |
| (Beautiful impressionistic oil painting) of a woman in the 1870s in france during the french revolution holding a three - pronged simple pitchfork and leading a rabble of shouting people to the bastille. She is clothed in french peasant clothing of the early 19th century and has a determined look on her face. | 33198401.jpeg |
| High quality photo in side view of a woman looking at the viewer standing on top of a hill in the windy scottish highlands wearing a tight full sleeve figure hugging turtle neck shirt and white dress pants that are also form fitted. | 33198417.jpeg |
| Misty Valley a close - up of a mysterious woman one half of her body and face is beautiful angelic appearance blue eyes white hair the other half of her body and face is demonic coal - black skin red eye and hair are burning with fire demonic horn on her back are wings corresponding to each part of the body. Dark atmosphere cinematic masterpiece clear focus. | 33198463.jpeg |
| High res waist up portrait photo of a woman with glittery eye - shadow and clear glossy lip - gloss. She is looking at the viewer with her mouth closed. She is wearing a thin string - like black choker and hoop earrings. In the background is a nightclub scene out of focus. | 33198688.jpeg |
| Abstract Surrealist cubist painting depicting a woman with green eyes and a red apple on her head looking at herself in the mirror. | 33198482.jpeg |
| High quality passport photo of a woman Neytiri with blue skin wearing a suit and tie looking directly at the camera with her mouth closed and a neutral expression. She is also wearing a delicate gold chain and some understated diamond earrings. | 33199007.jpeg |
| Instagram selfie of a nordic viking warrior princess with black warpaint under her eyes. There is a beautiful norwegian fjord in the background and other viking troops walking past her. She looks fierce and beautiful yet calm. | 33198486.jpeg |
| Anime screencap featuring a Minimalist and abstract side profile of a robot woman with a universe sprouting from her mind as she opens her eyes and attains consciousness for the first time in the dark. She has beautiful and intelligent bright orange eyes brimming with knowledge and benevolence. Mostly in hues of blue and green. | 33198556.jpeg |
| High quality portrait photo of an elven queen with pointy ears in the forrest wearing a laurel wreath. She is gazing at the viewer in an aloof and neutral manner. Voluminous rays of sunlight illuminating the forrest patch behind her. | 33198602.jpeg |
| Anime screencap featuring a Minimalist and abstract side profile of a robot woman with a universe sprouting from her mind as she opens her eyes and attains consciousness for the first time in the dark. She has beautiful and intelligent bright orange eyes brimming with knowledge and benevolence. Mostly in hues of blue and green. | 33198627.jpeg |
| A twitch streamer woman talking to the camera on her streaming setup sitting in a cute neon lit pink streamer room and wearing bunny ears. The orange neon sign in the background reads "CelebTV". | 33198751.jpeg |
| High quality passport photo of a woman wearing a suit and tie looking directly at the camera with her mouth closed and a neutral expression. She is also wearing a delicate gold chain and some understated diamond earrings. | 33243202.jpeg |
Stable Diffusion V1 5
Openrail
稳定扩散是一种潜在的文本到图像扩散模型，能够根据任何文本输入生成逼真的图像。
图像生成
S
stable-diffusion-v1-5
3.7M
518
Stable Diffusion Inpainting
Openrail
基于稳定扩散的文本到图像生成模型，具备图像修复能力
图像生成
S
stable-diffusion-v1-5
3.3M
56
Stable Diffusion Xl Base 1.0
SDXL 1.0是基于扩散的文本生成图像模型，采用专家集成的潜在扩散流程，支持高分辨率图像生成
图像生成
S
stabilityai
2.4M
6,545
Stable Diffusion V1 4
Openrail
稳定扩散是一种潜在文本到图像扩散模型，能够根据任意文本输入生成逼真图像。
图像生成
S
CompVis
1.7M
6,778
Stable Diffusion Xl Refiner 1.0
SD-XL 1.0优化器模型是Stability AI开发的图像生成模型，专为提升SDXL基础模型生成的图像质量而设计，特别擅长最终去噪步骤处理。
图像生成
S
stabilityai
1.1M
1,882
Stable Diffusion 2 1
基于扩散的文本生成图像模型，支持通过文本提示生成和修改图像
图像生成
S
stabilityai
948.75k
3,966
Stable Diffusion Xl 1.0 Inpainting 0.1
基于Stable Diffusion XL的潜在文本到图像扩散模型，具备通过遮罩进行图像修复的功能
图像生成
S
diffusers
673.14k
334
Stable Diffusion 2 Base
基于扩散的文生图模型，可根据文本提示生成高质量图像
图像生成
S
stabilityai
613.60k
349
Playground V2.5 1024px Aesthetic
其他
开源文生图模型，能生成1024x1024分辨率及多种纵横比的美学图像，在美学质量上处于开源领域领先地位。
图像生成
P
playgroundai
554.94k
723
Sd Turbo
SD-Turbo是一款高速文本生成图像模型，仅需单次网络推理即可根据文本提示生成逼真图像。该模型作为研究原型发布，旨在探索小型蒸馏文本生成图像模型。
图像生成
S
stabilityai
502.82k
380
精选推荐AI模型
Llama 3 Typhoon V1.5x 8b Instruct
专为泰语设计的80亿参数指令模型，性能媲美GPT-3.5-turbo，优化了应用场景、检索增强生成、受限生成和推理任务
大型语言模型
Transformers 支持多种语言
Transformers 支持多种语言L
scb10x
3,269
16
Cadet Tiny
Openrail
Cadet-Tiny是一个基于SODA数据集训练的超小型对话模型，专为边缘设备推理设计，体积仅为Cosmo-3B模型的2%左右。
对话系统 Transformers 英语
Transformers 英语C
ToddGoldfarb
2,691
6
Roberta Base Chinese Extractive Qa
基于RoBERTa架构的中文抽取式问答模型，适用于从给定文本中提取答案的任务。
问答系统 中文
R
uer
2,694
98