Files
hao-ai-lab-FastVideo/docs/getting_started/installation.md
T

68 lines
1.9 KiB
Markdown

# 🔧 Installation
FastVideo supports the following hardware platforms:
- [NVIDIA CUDA](installation/gpu.md)
- [NVIDIA DGX Spark / GB10 (ARM64 + CUDA 13)](installation/spark.md)
- [Apple silicon](installation/mps.md)
## Quick Installation
### Using uv (recommended)
Use uv as the default environment manager for faster and more stable installs.
The commands below target NVIDIA CUDA 12; use `UV_TORCH_BACKEND=cu130` on
CUDA 13. Apple silicon users should follow the [MPS guide](installation/mps.md).
```bash
# Create and activate a new uv environment
uv venv --python 3.12 --seed
source .venv/bin/activate
UV_TORCH_BACKEND=cu126 uv pip install fastvideo
```
### Using Conda (alternative)
```bash
# Create and activate a new conda environment
conda create -n fastvideo python=3.12 -y
conda activate fastvideo
UV_TORCH_BACKEND=cu126 uv pip install fastvideo
```
### From source
FastVideo pins PyTorch 2.12.0. Select its CUDA build explicitly with
`UV_TORCH_BACKEND=cu126` for CUDA 12 or `UV_TORCH_BACKEND=cu130` for CUDA 13.
```bash
git clone https://github.com/hao-ai-lab/FastVideo.git
cd FastVideo
UV_TORCH_BACKEND=cu126 uv pip install -e .
# optional: install flash-attn
uv pip install flash-attn --no-build-isolation -v
```
Alternative with Conda environment (still drives installs through `uv`):
```bash
UV_TORCH_BACKEND=cu126 uv pip install -e .
uv pip install flash-attn --no-build-isolation -v
```
## Hardware Requirements
- **NVIDIA GPUs**: CUDA 12.6+ with compute capability 7.0+
- **Apple Silicon**: macOS 14.0+ with M1/M2/M3/M4 chips
- **CPU**: x86_64 architecture (for CPU-only inference)
## Next Steps
- [Quick Start Guide](quick_start.md) - Get started with your first video generation
- [Configuration](../inference/configuration.md) - Learn about configuration options
- [Examples](../inference/examples/examples_inference_index.md) - Explore example scripts and notebooks