HiDream-I1-Fast-nf4 Open-source Image Generation Model - Run with Low Video Memory and Quickly Produce High-quality Images

Hidream I1 Fast Nf4

Developed by azaneko

HiDream-I1 is an open-source image generation foundation model with 17 billion parameters. The 4-bit quantized version can run on 16GB VRAM, enabling fast and high-quality image generation.

Image Generation Open Source License:MIT #4-bit quantization #Fast image generation #Low VRAM requirement

Downloads 19.22k

Release Time : 4/8/2025

Model Overview

HiDream-I1 is a high-performance text-to-image generation model. After 4-bit quantization, its VRAM requirement is reduced, allowing it to run on consumer-grade GPUs. The fast version employs distillation techniques to reduce inference steps, making it suitable for scenarios requiring rapid image generation.

Model Features

Efficient Quantization

4-bit quantization technology enables the model to run on consumer-grade GPUs with 16GB VRAM

Fast Inference

The distilled model requires only 16 inference steps to generate high-quality images

High-Quality Generation

The 17 billion parameter foundation model delivers state-of-the-art image generation quality

Multi-GPU Support

Supports NVIDIA GPUs with Ampere architecture and above (e.g., RTX 3090/4090)

Model Capabilities

Text-to-image generation

High-quality image synthesis

Fast image generation

Use Cases

Creative Design

Concept Art Creation

Quickly generate concept art images based on text descriptions

Can produce high-quality concept art in seconds

Advertising Material Generation

Automatically generate product promotional images

Rapidly produce diverse advertising materials

Content Creation

Social Media Content

Generate accompanying images for social media

Quickly respond to image demands for trending topics

🚀 HiDream-I1 4Bit Quantized Model

This repository offers a 4-bit quantized version of HiDream-I1, enabling the full model to run with less than 16GB of VRAM.

The original repository can be found here.

HiDream-I1 is a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds.

🚀 Quick Start

Simply run:

pip install hdi1 --no-build-isolation

⚠️ Important Note

It's recommended that you start a new python environment for this package to avoid dependency conflicts.
To do that, you can use conda create -n hdi1 python=3.12 and then conda activate hdi1.
Or you can use python3 -m venv venv and then source venv/bin/activate on Linux or venv\Scripts\activate on Windows.

Command Line Interface

Then you can run the module to generate images:

python -m hdi1 "A cat holding a sign that says 'hello world'"

# or you can specify the model
python -m hdi1 "A cat holding a sign that says 'hello world'" -m fast

⚠️ Important Note

The inference script will try to automatically download meta-llama/Llama-3.1-8B-Instruct model files. You need to agree to the license of the Llama model on your HuggingFace account and login using huggingface-cli login in order to use the automatic downloader.

Web Dashboard

We also provide a web dashboard for interactive image generation. You can start it by running:

python -m hdi1.web

✨ Features

We offer both the full version and distilled models. The parameter size are the same, so they require the same amount of GPU memory to run. However, the distilled models are faster because of reduced number of inference steps.

Name	Min VRAM	Steps	HuggingFace
HiDream-I1-Full	16 GB	50	🤗 Original / NF4
HiDream-I1-Dev	16 GB	28	🤗 Original / NF4
HiDream-I1-Fast	16 GB	16	🤗 Original / NF4

📦 Installation

The installation command is as follows:

pip install hdi1 --no-build-isolation

📚 Documentation

Hardware Requirements

GPU Architecture: NVIDIA >= Ampere (e.g. A100, H100, A40, RTX 3090, RTX 4090)
GPU RAM: >= 16 GB
CPU RAM: >= 16 GB

📄 License

The code in this repository and the HiDream-I1 models are licensed under MIT License.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご