🎨 SDXL Model Converter

Convert SDXL checkpoints to Diffusers format (FP16, CPU-only).

📥 Input Sources Supported:

Local model files (.safetensors, .ckpt)
Direct URLs to model files
Hugging Face model repositories (e.g., 'my-org/my-model' or 'my-org/my-model/file.safetensors')

ℹ️ Important Notes:

This tool runs on CPU, conversion might be slower than on GPU.
For Hugging Face uploads, you need a WRITE token (not a read token).
Get your HF token here: https://huggingface.co/settings/tokens

💾 Memory Usage:

This space is configured for FP16 precision to reduce memory usage.
Close other applications during conversion.
For large models, ensure you have at least 16GB of RAM.

💻 Source Code:

GitHub Repository

🙏 Support:

If you're interested in funding more projects: Ko-fi

SDXL Checkpoint (Path, URL, or HF Repo)

Reference Diffusers Model (Optional)

Output Path (Diffusers Format)

Hugging Face Token

Organization Name (Optional)

Model Name

Make Repository Private