English注册

Alissonerdx/TryAnything

🤗 Hugging Face 来源text-to-imageapache-2.0389 MBother✓ 17 个校验和今天更新
一条命令提交

在你的模型文件夹旁边运行它。它会制作种子、将文件与 Hugging Face 比对，然后提交。你只需开始做种，并粘贴你账户中的密钥。它只读取你的文件，绝不修改。如果愿意，可以先阅读脚本。

curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo Alissonerdx/TryAnything ./model-folder
需要做种者 →

Try Anything

Model description

Model Type: LoRA
Trained On: Flux Kontext (via AI Toolkit)
Best Results With: Flux Kontext / Flux Dev (untested but likely compatible)
Trigger Word: `try on this outfit`
Purpose: Virtual clothing try-on for realistic outfit visualization

Workflow: Available on Patreon

🧠 What is TryAnything?

TryAnything is a virtual try-on LoRA designed to transfer clothing from product images onto people. It enables realistic visualization of how outfits would look on different individuals while preserving identity.

The model was trained using 240 real-person images from VTON datasets, focusing primarily on upper body garments with the ability to handle full outfit combinations.

🎯 What TryAnything Does

TryAnything transfers clothing details including:

  • Upper body garments (shirts, blouses, jackets, tops)
  • Full outfit combinations (top + bottom)
  • Clothing fit and draping
  • Fabric textures and patterns
  • Natural clothing appearance on the target person

🧪 Training Info

  • Trainer: AI Toolkit
  • Steps: 6,000
  • Dataset: 240 real-person VTON images
  • LoRA Rank: 32
  • Resolution: 1024px
  • Base Model: Flux Kontext
  • Focus: Upper body garments with full outfit capability

⚙️ Usage Guide

Recommended Settings:

  • LoRA Weight: 1.0 (adjust as needed)
  • Trigger: `woman try on this outfit` or `man try on this outfit`
  • Best Practice: Use clothing images without people wearing them
  • Can work: With people wearing clothes, but not recommended

Example Prompts:

try on this outfit

try on this outfit, man

try on this outfit, man, copy t-shirt and shorts

try on this outfit, blouse with short sleeves

⚠️ Important Notes - Version 1 Limitations

Current Limitations:

  • V1 Status: This is the first version with room for improvement
  • Head Proportions: May have issues with head-to-body proportions (most common problem)
  • Training Focus: Primarily trained on upper body garments
  • Bottom Wear: Limited training on lower body clothing

Best Results:

  • Use standalone clothing images (not worn by people)
  • Specify clothing details in prompts when needed
  • Include "anatomically correct proportions" in prompts

Future V2 Plans:

  • Enhanced training with underwear and foundation garments
  • Improved lower body clothing handling
  • Better proportion consistency

📥 How to Use

  1. Load the LoRA in your workflow
  2. Use trigger: `woman try on this outfit` or `man try on this outfit`
  3. Provide clear clothing reference images (preferably unworn)
  4. Include clothing details in prompt if needed
  5. Add "anatomically correct proportions" for better results

🔬 Compatibility

  • Primary: Flux Kontext
  • Likely Compatible: Flux Dev (untested)
  • Weight: Start with 1.0, adjust up/down as needed

Feedback Welcome! This is V1 - share your results and suggestions for improvements in V2!

Trigger words

You should use try on this outfit to trigger the image generation.

You should use man try on this outfit to trigger the image generation.

You should use woman try on this outfit to trigger the image generation.

Download model

Download them in the Files & versions tab.