Autoxhs is an open-source tool designed to streamline and automate the process of generating and publishing content on Xiaohongshu (Little Red Book). By leveraging the power of OpenAI's API, Autoxhs provides a comprehensive solution for creating engaging images, titles, content, and tags with ease.
🚀 Try Our App Live on Streamlit Cloud
Table of Contents
Features
- Automated Content Creation: Leverages OpenAI's API for generating engaging images, titles, content, and tags, streamlining content creation for Xiaohongshu.
- LangGPT Prompt Generation & Langchian Framework: Utilizes the LangGPT project methodology and Langchian framework for creating structured, high-quality prompts and developing language model-powered applications.
- Customizable & User-Friendly: Offers prompt customization and model selection options within a Streamlit-based interface, ensuring a user-friendly experience.
- Efficient Post Generation: Supports both topic and image-based post generation. Users can input a topic or upload photos, and Autoxhs automatically crafts a complete, engaging post, simplifying the content creation process.
Demo
Quick demonstrations of Autoxhs (videos accelerated for brevity):
- Theme-Based Post Generation: Users simply input a topic, and Autoxhs automatically crafts a complete post, including both text content and matching images.
- Image-Based Post Generation: Users can upload their own photos, and Autoxhs will generate rich and engaging post content based on these images.
Getting Started
- To install Autoxhs, follow these steps:
# Create a new conda environment with Python 3.9
conda create -n Autoxhs python=3.9
# Activate the conda environment
conda activate Autoxhs
# Clone the Autoxhs repository to your local machine
git clone https://github.com/Gikiman/Autoxhs.git
# Navigate to the Autoxhs project directory
cd Autoxhs
# Install dependencies from requirements.txt
pip install -r requirements.txt
# If you are using a Linux system, you'll need to install the corresponding dependencies
# xargs -a packages.txt sudo apt-get install -y
- After installing, you can launch the application simply by running:
streamlit run Autoxhs.py
Note: A high-speed internet connection is required, and users in China should use a proxy.
Configuration
- Prompt Customization: Modify generation prompts in the
data\prompt
folder using the LangGPT methodology. - Model Selection: Default models are
gpt-4-0125-preview
for text anddall-e-3
for images. These can be selected in the app.