# MindIE SD

MindIE SD is an Ascend-focused inference acceleration toolkit for Stable Diffusion and related multimodal generation workloads. It provides Ascend-oriented fused operators, compilation paths, caching strategies, quantization features, and multi-card execution capabilities for production-oriented inference workflows.

:maxdepth: 2
:caption: Getting Started

quick_start
installing_guide
architecture
environment_variable_configuration

:maxdepth: 2
:caption: Feature Guides

features/cache
features/compilation
features/DyEPLB
features/graphics_memory_optimization
features/others
features/parallelism
features/sparse_quantization
features/supported_matrix

:maxdepth: 2
:caption: Developer Docs

developer_guide/build_guide
developer_guide/test
developer_guide/tooling

:maxdepth: 2
:caption: Appendix

appendix/environment_variable
appendix/file_directory_permissions_description

:maxdepth: 1
:caption: Community

community/governance