Hanni Makina Flux

🚀 Hanni Makina FLUX

Hanni Makina FLUX 是一款基於 Stable Diffusion 的 LoRA 模型，可用於文本到圖像的生成。它以特定的基礎模型為依託，能根據輸入的文本生成具有特定風格和特徵的圖像，適用於對亞洲年輕女性形象的圖像創作。

🚀 快速開始

模型下載

此模型的權重以 Safetensors 格式提供。你可以在 Files & versions 標籤頁 中下載。

使用 🧨 diffusers 庫

以下是使用該模型的示例代碼：

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/hanni-makina-flux', weight_name='makinaflux_hanni_v1.0.safetensors')
image = pipeline('a petite young asian woman  posing in a stylish and coordinated outfit. She is wearing a white sleeveless crop top paired with a matching white mini skirt, creating a chic and fashionable look. The top features a black and white tie with "Chanel" branding, adding a luxurious touch to the ensemble. Her long hair is styled in two high pigtails, decorated with accessories, giving her a youthful and trendy appearance. The background is minimalistic, with gray walls, keeping the focus on her outfit and pose. Her expression is neutral and poised, suggesting a confident and composed attitude.. ,, . , looking at camera').images[0]

更多詳細信息，包括 LoRA 的加權、合併和融合，請查看 diffusers 中加載 LoRA 的文檔。

✨ 主要特性

圖像生成示例

• 示例一：輸入文本描述為 “This image showcases a young woman in a minimalistic, high - fashion pose. She is turned slightly to her right, with her face and upper body in focus. Her dark hair is cut in a sleek, straight bob with blunt bangs that frame her face, giving her a modern, edgy look. A few strands of hair are gently blown out of place, adding a dynamic and natural feel to the composition... ”，輸出圖像為 32022667.jpeg。
• 示例二：輸入 “bright photo of beautiful east asian girl wearing bohemian dress, inside a rustic house with windows, necklace, dslr, studio lighting, high quality, film grain, light reflections, blood vessels, pale skin, detailed skin”，輸出圖像為 32022672.jpeg。
• 示例三：輸入 “The image shows a petite young asian woman standing in front of a pink background with text on it, possibly part of a poster or event backdrop. She is dressed in a unique and playful outfit that includes a light - colored oversized hoodie or sweatshirt and a denim hat with embellishments... ”，輸出圖像為 32026248.jpeg。
• 示例四：輸入 “he image features a petite young asian woman on stage, wearing a casual yet stylish outfit. She is dressed in a white cropped top with a denim jacket and high - waisted denim shorts, creating a coordinated look. She has long, straight black hair and is wearing glasses... ”，輸出圖像為 32026322.jpeg。
• 示例五：輸入 “a petite young asian woman posing in a stylish and coordinated outfit. She is wearing a white sleeveless crop top paired with a matching white mini skirt, creating a chic and fashionable look. The top features a black and white tie with "Chanel" branding... ”，輸出圖像為 32026337.jpeg。

模型使用參數

• 採樣方法：使用 LMS/IPNDM/IPNDM_V 可獲得最佳皮膚紋理，使用 EULER 可獲得更光滑的皮膚效果。
• 權重：建議權重範圍為 0.8 - 1.0。
• 採樣步數：特寫 20 - 25 步，半身/上半身 25 - 30 步，全身 30 - 35 步。
• HiRes Fix：特寫去噪強度 0.15 - 0.2，半身/上半身 0.2 - 0.25，全身 0.25 - 0.3。
• Adetailer：僅在 3/4 身到全身圖像時需要，內繪去噪強度 0.2 - 0.3。

📄 許可證

本模型採用 [bespoke - lora - trained - license](https://multimodal.art/civitai - licenses?allowNoCredit=True&allowCommercialUse=RentCivit&allowDerivatives=False&allowDifferentLicense=True) 許可證。

🔗 支持與教程

• 支持作者：[https://ko - fi.com/makina69](https://ko - fi.com/makina69)
• Flux 教程：Makina's Lazy FLUX.1d Starter Guide (한국어 & 日本語) | Civitai