v1.0.0 - Initial Release

🎉 Initial Release - HunyuanImage-3.0 ComfyUI Nodes

Professional ComfyUI custom nodes for Tencent HunyuanImage-3.0, the powerful 80B parameter native multimodal image generation model.

✨ Features

• Multiple Loading Modes: Full BF16, NF4 Quantized, Single GPU, Multi-GPU
• Smart Memory Management: Automatic VRAM tracking and cleanup
• High-Quality Generation: Standard (<2MP) and Large (2MP-8MP+) image support
• Advanced Prompting: Official HunyuanImage-3.0 prompt enhancement with OpenAI-compatible LLM APIs
• Professional Resolution Control: Organized presets with megapixel indicators

📋 Requirements

• ComfyUI installed and working
• NVIDIA GPU: 24GB+ VRAM for NF4, 80GB+ for BF16
• Python 3.10+, PyTorch 2.7+, CUDA 12.8+

🚀 Installation

cd ComfyUI/custom_nodes
git clone https://github.com/EricRollei/Comfy_HunyuanImage3.git
cd Comfy_HunyuanImage3
pip install -r requirements.txt

See README.md for full documentation.

🙏 Credits

This project integrates the HunyuanImage-3.0 model developed by Tencent Hunyuan Team under Apache 2.0 license.