Tencent-Hunyuan/HunyuanImage-3.0

中文文档

🎨 HunyuanImage-3.0: A Powerful Native Multimodal Model for Image Generation

👏 Join our WeChat and Discord | 💻 Official website(官网) Try our model!&nbsp&nbsp

🔥🔥🔥 News

• January 26, 2026: 🚀 HunyuanImage-3.0-Instruct-Distil - Distilled checkpoint for efficient deployment (8 steps sampling recommended).
• January 26, 2026: 🎉 HunyuanImage-3.0-Instruct - Release of Instruct (with reasoning) for intelligent prompt enhancement and Image-to-Image generation for creative editing.
• October 30, 2025: 🚀 HunyuanImage-3.0 vLLM Acceleration - Significantly faster inference with vLLM support.
• September 28, 2025: 📖 HunyuanImage-3.0 Technical Report - Comprehensive technical documentation now available.
• September 28, 2025: 🎉 HunyuanImage-3.0 Open Source - Inference code and model weights publicly available.

🧩 Community Contributions

If you develop/use HunyuanImage-3.0 in your projects, welcome to let us know.

📑 Open-source Plan

• HunyuanImage-3.0 (Image Generation Model)
• [x] Inference
• [x] HunyuanImage-3.0 Checkpoints
• [x] HunyuanImage-3.0-Instruct Checkpoints (with reasoning)
• [x] vLLM Support
• [x] Distilled Checkpoints
• [x] Image-to-Image Generation
• [ ] Multi-turn Interaction

🗂️ Contents

• 🔥🔥🔥 News
• 🧩 Community Contributions
• 📑 Open-source Plan
• 📖 Introduction
• ✨ Key Features
• 🚀 Usage
• 📦 Environment Setup
• 📥 Install Dependencies
• HunyuanImage-3.0-Instruct
• 🔥 Quick Start with Transformers
• 1️⃣ Download model weights
• 2️⃣ Run with Transformers
• 🏠 Local Installation & Usage
• 1️⃣ Clone the Repository
• 2️⃣ Download Model Weights
• 3️⃣ Run the Demo
• 4️⃣ Command Line Arguments
• 5️⃣ For fewer Sampling Steps
• HunyuanImage-3.0 (Text-to-image)
• 🔥 Quick Start with Transformers
• 1️⃣ Download model weights
• 2️⃣ Run with Transformers
• 🏠 Local Installation & Usage
• 1️⃣ Clone the Repository
• 2️⃣ Download Model Weights
• 3️⃣ Run the Demo
• 4️⃣ Command Line Arguments
• 🎨 Interactive Gradio Demo
• 1️⃣ Install Gradio
• 2️⃣ Configure Environment
• 3️⃣ Launch the Web Interface
• 4️⃣ Access the Interface
• 🧱 Models Cards
• 📊 Evaluation
• Evaluation of HunyuanImage-3.0-Instruct
• Evaluation of HunyuanImage-3.0 (Text-to-Image)
• 🖼️ Showcase
• Showcases of HunyuanImage-3.0-Instruct
• 📚 Citation
• 🙏 Acknowledgements
• 🌟🚀 Github Star History

📖 Introduction

HunyuanImage-3.0 is a groundbreaking native multimodal model that unifies multimodal understanding and generation within an autoregressive framework. Our text-to-image and image-to-image model achieves performance comparable to or surpassing leading closed-source models.

✨ Key Features

• 🧠 Unified Multimodal Architecture: Moving beyond the prevalent DiT-based architectures, HunyuanImage-3.0 employs a unified autoregressive framework. This design enables a more direct and integrated modeling of text and image modalities, leading to surprisingly effective and contextually rich image generation.
• 🏆 The Largest Image Generation MoE Model: This is the largest open-source image generation Mixture of Experts (MoE) model to date. It features 64 experts and a total of 80 billion parameters, with 13 billion activated per token, significantly enhancing its capacity and performance.
• 🎨 Superior Image Generation Performance: Through rigorous dataset curation and advanced reinforcement learning post-training, we've achieved an optimal balance between semantic accuracy and visual excellence. The model demonstrates exceptional prompt adherence while delivering photorealistic imagery with stunning aesthetic quality and fine-grained details.
• 💭 Intelligent Image Understanding and World-Knowledge Reasoning: The unified multimodal architecture endows HunyuanImage-3.0 with powerful reasoning capabilities. It under stands user's input image, and leverages its extensive world knowledge to intelligently interpret user intent, automatically elaborating on sparse prompts with contextually appropriate details to produce superior, more complete visual outputs.

🚀 Usage

📦 Environment Setup

• 🐍 Python: 3.12+ (recommended and tested)
• ⚡ CUDA: 12.8

📥 Install Dependencies

# 1. First install PyTorch (CUDA 12.8 Version)
pip install torch==2.8.0 torchvision==0.23.0 torchaudio==2.8.0 --index-url https://download.pytorch.org/whl/cu128
2. Install tencentcloud-sdk for Prompt Enhancement (PE) only for HunyuanImage-3.0 not HunyuanImage-3.0-Instruct
pip install -i https://mirrors.tencent.com/pypi/simple/ --upgrade tencentcloud-sdk-python
3. Then install other dependencies
pip install -r requirements.txt

For up to 3x faster inference, install these optimizations:

# FlashInfer for optimized moe inference. v0.5.0 is tested.
pip install flashinfer-python==0.5.0

💡Installation Tips: It is critical that the CUDA version used by PyTorch matches the system's CUDA version.

FlashInfer relies on this compatibility when compiling kernels at runtime.

GCC version >=9 is recommended for compiling FlashAttention and FlashInfer.

⚡ Performance Tips: These optimizations can significantly speed up your inference!

💡Notation: When FlashInfer is enabled, the first inference may be slower (about 10 minutes) due to kernel compilation. Subsequent inferences on the same machine will be much faster.

HunyuanImage-3.0-Instruct (Instruction reasoning and Image-to-image generation, including editing and multi-image fusion)

🔥 Quick Start with Transformers

1️⃣ Download model weights

# Download from HuggingFace and rename the directory.
Notice that the directory name should not contain dots, which may cause issues when loading using Transformers.
hf download tencent/HunyuanImage-3.0-Instruct --local-dir ./HunyuanImage-3-Instruct

2️⃣ Run with Transformers

from transformers import AutoModelForCausalLM
Load the model
model_id = "./HunyuanImage-3-Instruct"
Currently we can not load the model using HF model_id tencent/HunyuanImage-3.0-Instruct directly 
due to the dot in the name.
kwargs = dict(
    attn_implementation="sdpa", 
    trust_remote_code=True,
    torch_dtype="auto",
    device_map="auto",
    moe_impl="eager",   # Use "flashinfer" if FlashInfer is installed
    moe_drop_tokens=True,
)
model = AutoModelForCausalLM.from_pretrained(model_id, kwargs)
model.load_tokenizer(model_id)

Image-to-Image generation (TI2I)
prompt = "基于图一的logo，参考图二中冰箱贴的材质，制作一个新的冰箱贴"
input_img1 = "./assets/demo_instruct_imgs/input_1_0.png"
input_img2 = "./assets/demo_instruct_imgs/input_1_1.png"
imgs_input = [input_img1, input_img2]
cot_text, samples = model.generate_image(
    prompt=prompt,
    image=imgs_input,
    seed=42,
    image_size="auto",
    use_system_prompt="en_unified",
    bot_task="think_recaption",  # Use "think_recaption" for reasoning and enhancement
    infer_align_image_size=True,  # Align output image size to input image size
    diff_infer_steps=50, 
    verbose=2
)
Save the generated image
samples[0].save("image_edit.png")

git clone https://github.com/Tencent-Hunyuan/HunyuanImage-3.0.git
cd HunyuanImage-3.0/

# Download from HuggingFace
hf download tencent/HunyuanImage-3.0-Instruct --local-dir ./HunyuanImage-3-Instruct

export MODEL_PATH="./HunyuanImage-3-Instruct"
bash run_demo_instruct.sh

We recommend using the model HunyuanImage-3.0-Instruct-Distil with --diff-infer-steps 8, while keeping all other recommended parameter values unchanged.

# Download HunyuanImage-3.0-Instruct-Distil from HuggingFace
hf download tencent/HunyuanImage-3.0-Instruct-Distil --local-dir ./HunyuanImage-3-Instruct-Distil
Run the demo with 8 steps to samples
export MODEL_PATH="./HunyuanImage-3-Instruct-Distil"
bash run_demo_instruct_Distil.sh

# Download from HuggingFace and rename the directory.
Notice that the directory name should not contain dots, which may cause issues when loading using Transformers.
hf download tencent/HunyuanImage-3.0 --local-dir ./HunyuanImage-3

from transformers import AutoModelForCausalLM
Load the model
model_id = "./HunyuanImage-3"
Currently we can not load the model using HF model_id tencent/HunyuanImage-3.0 directly 
due to the dot in the name.
kwargs = dict(
    attn_implementation="sdpa",     # Use "flash_attention_2" if FlashAttention is installed
    trust_remote_code=True,
    torch_dtype="auto",
    device_map="auto",
    moe_impl="eager",   # Use "flashinfer" if FlashInfer is installed
)
model = AutoModelForCausalLM.from_pretrained(model_id, kwargs)
model.load_tokenizer(model_id)
generate the image
prompt = "A brown and white dog is running on the grass"
image = model.generate_image(prompt=prompt, stream=True)
image.save("image.png")

git clone https://github.com/Tencent-Hunyuan/HunyuanImage-3.0.git
cd HunyuanImage-3.0/

# Download from HuggingFace
hf download tencent/HunyuanImage-3.0 --local-dir ./HunyuanImage-3

# Without PE
export MODEL_PATH="./HunyuanImage-3"
python3 run_image_gen.py \
    --model-id $MODEL_PATH \
    --verbose 1 \
    --prompt "A brown and white dog is running on the grass" \
    --bot-task image \
    --image-size "1024x1024" \
    --save ./image.png \
    --moe-impl flashinfer
With PE
export DEEPSEEK_KEY_ID="your_deepseek_key_id"
export DEEPSEEK_KEY_SECRET="your_deepseek_key_secret"
export MODEL_PATH="./HunyuanImage-3"
python3 run_image_gen.py \
    --model-id $MODEL_PATH \
    --verbose 1 \
    --prompt "A brown and white dog is running on the grass" \
    --bot-task image \
    --image-size "1024x1024" \
    --save ./image.png \
    --moe-impl flashinfer \
    --rewrite 1

pip install gradio>=4.21.0

# Set your model path
export MODEL_ID="path/to/your/model"
Optional: Configure GPU usage (default: 0,1,2,3)
export GPUS="0,1,2,3"
Optional: Configure host and port (default: 0.0.0.0:443)
export HOST="0.0.0.0"
export PORT="443"

sh run_app.sh

# Use both optimizations for maximum performance
sh run_app.sh --moe-impl flashinfer --attn-impl flash_attention_2

🧱 Models Cards

| Model | Params | Download | Recommended VRAM | Supported | |---------------------------| --- | --- | --- | --- | | HunyuanImage-3.0 | 80B total (13B active) | HuggingFace | ≥ 3 × 80 GB | ✅ Text-to-Image | HunyuanImage-3.0-Instruct | 80B total (13B active) | HuggingFace | ≥ 8 × 80 GB | ✅ Text-to-Image
✅ Text-Image-to-Image
✅ Prompt Self-Rewrite
✅ CoT Think | HunyuanImage-3.0-Instruct-Distil | 80B total (13B active) | HuggingFace | ≥ 8 × 80 GB |✅ Text-to-Image
✅ Text-Image-to-Image
✅ Prompt Self-Rewrite
✅ CoT Think
✅ Fewer sampling steps (8 steps recommended)

Notes:

• Install performance extras (FlashAttention, FlashInfer) for faster inference.
• Multi‑GPU inference is recommended for the Base model.

📊 Evaluation

Evaluation of HunyuanImage-3.0-Instruct

• 👥 GSB (Human Evaluation)

Evaluation of HunyuanImage-3.0 (Text-to-Image)

• 🤖 SSAE (Machine Evaluation)

• 👥 GSB (Human Evaluation)

🖼️ Showcase

Our model can follow complex instructions to generate high‑quality, creative images.

For text-to-image showcases in HunyuanImage-3.0, click the following links:

• HunyuanImage-3.0

Showcases of HunyuanImage-3.0-Instruct

HunyuanImage-3.0-Instruct demonstrates powerful capabilities in intelligent image generation and editing. The following showcases highlight its core features:

• 🧠 Intelligent Visual Understanding and Reasoning (CoT Think): The model performs structured thinking to analyze user's input image and prompt, expand user's intent and editing tasks into a stucture, comprehnsive instructions, and leading to a better image generation and editing performance.

• ✏️ Prompt Self-Rewrite: Automatically enhances sparse or vague prompts into professional-grade, detail-rich descriptions that capture the user's intent more accurately.
• 🎨 Text-to-Image (T2I): Generates high-quality images from text prompts with exceptional prompt adherence and photorealistic quality.
• 🖼️ Image-to-Image (TI2I): Supports creative image editing, including adding elements, removing objects, modifying styles, and seamless background replacement while preserving key visual elements.
• 🔀 Multi-Image Fusion: Intelligently combines multiple reference images (up to 3 inputs) to create coherent composite images that integrate visual elements from different sources.

Showcase 1: Detailed Thought and Reasoning Process

Showcase 2: Creative T2I Generation with Complex Scene Understanding

Prompt: 3D 毛绒质感拟人化马，暖棕浅