Installation¶

PyPI: pip install torch-sla — simplest installation
GitHub: Clone and install for development
Optional backends: cuDSS, Eigen for enhanced performance

Using pip¶

To install the latest release:

pip install torch-sla

Or install from GitHub for the latest development version:

pip install git+https://github.com/walkerchi/torch-sla.git

Optional Dependencies¶

For additional backends and features:

# With cuDSS support (requires CUDA 12+)
pip install torch-sla[cuda]

# Full installation with all optional dependencies
pip install torch-sla[all]

# For development
pip install torch-sla[dev]

cuDSS Now on PyPI!

CUDA backends use nvmath-python (for cuDSS) and cupy-cuda12x (for CuPy). Installing torch-sla[cuda] will automatically install them.

Backend Requirements¶

Backend	Installation	Notes
`scipy`	`pip install scipy`	Default, always available
`pytorch`	Included with PyTorch	Native CG/BiCGStab solvers
`cupy`	`pip install cupy-cuda12x`	GPU direct + iterative solvers via cupyx.scipy
`cudss`	`pip install nvmath-python[cu12]`	Best for medium-scale GPU problems (10K-2M DOF)