返回 Pixelle-Video
installation.md
根目录 / docs / en / getting-started / installation.md
1 # Installation
2
3 This page will guide you through installing Pixelle-Video.
4
5 ---
6
7 ## System Requirements
8
9 ### Required
10
11 - **Python**: 3.10 or higher
12 - **Operating System**: Windows, macOS, or Linux
13 - **Package Manager**: uv (recommended) or pip
14
15 ### Optional
16
17 - **GPU**: NVIDIA GPU with 6GB+ VRAM recommended for local ComfyUI
18 - **Network**: Stable internet connection for LLM API and image generation services
19
20 ---
21
22 ## 🪟 Windows All-in-One Package (Recommended for Windows Users)
23
24 **No need to install Python, uv, or ffmpeg - ready to use out of the box!**
25
26 ### Download and Install
27
28 1. Visit [GitHub Releases](https://github.com/AIDC-AI/Pixelle-Video/releases/latest) to download the latest version
29 2. Download the latest Windows All-in-One Package and extract it to any directory
30 3. Double-click `start.bat` to launch the Web interface
31 4. Your browser will automatically open `http://localhost:8501`
32
33 !!! success "Installation Complete!"
34 The package includes all dependencies, no need to manually install any environment. On first use, you only need to configure API keys in "⚙️ System Configuration" to get started.
35
36 !!! tip "Next Steps"
37 After installation, check out the [Configuration Guide](configuration.md) to set up LLM and image generation services, then see [Quick Start](quick-start.md) to create your first video.
38
39 ---
40
41 ## Install from Source (For macOS / Linux Users or Users Who Need Customization)
42
43 ### Step 1: Clone the Repository
44
45 ```bash
46 git clone https://github.com/AIDC-AI/Pixelle-Video.git
47 cd Pixelle-Video
48 ```
49
50 ### Step 2: Install Dependencies
51
52 !!! tip "Recommended: Use uv"
53 This project uses `uv` as the package manager, which is faster and more reliable than traditional pip.
54
55 #### Using uv (Recommended)
56
57 ```bash
58 # Install uv if you haven't already
59 curl -LsSf https://astral.sh/uv/install.sh | sh
60
61 # Install project dependencies (uv will create a virtual environment automatically)
62 uv sync
63 ```
64
65 #### Using pip
66
67 ```bash
68 # Create virtual environment
69 python -m venv venv
70
71 # Activate virtual environment
72 # Windows:
73 venv\Scripts\activate
74 # macOS/Linux:
75 source venv/bin/activate
76
77 # Install dependencies
78 pip install -e .
79 ```
80
81 ---
82
83 ## Verify Installation
84
85 Run the following command to verify the installation:
86
87 ```bash
88 # Using uv
89 uv run streamlit run web/app.py
90
91 # Or using pip (activate virtual environment first)
92 streamlit run web/app.py
93 ```
94
95 Your browser should automatically open `http://localhost:8501` and display the Pixelle-Video web interface.
96
97 !!! success "Installation Successful!"
98 If you can see the web interface, the installation was successful! Next, check out the [Configuration Guide](configuration.md) to set up your services.
99
100 ---
101
102 ## Optional: Install ComfyUI (Local Deployment)
103
104 If you want to run image generation locally, you'll need to install ComfyUI:
105
106 ### Quick Install
107
108 ```bash
109 # Clone ComfyUI
110 git clone https://github.com/comfyanonymous/ComfyUI.git
111 cd ComfyUI
112
113 # Install dependencies
114 pip install -r requirements.txt
115 ```
116
117 ### Start ComfyUI
118
119 ```bash
120 python main.py
121 ```
122
123 ComfyUI runs on `http://127.0.0.1:8188` by default.
124
125 !!! info "ComfyUI Models"
126 ComfyUI requires downloading model files to work. Please refer to the [ComfyUI documentation](https://github.com/comfyanonymous/ComfyUI) for information on downloading and configuring models.
127
128 ---
129
130 ## Next Steps
131
132 - [Configuration](configuration.md) - Configure LLM and image generation services
133 - [Quick Start](quick-start.md) - Create your first video
134
135
135 lines MARKDOWN