Documentation History¶
This page records documentation changes by release.
For runtime, API, and compatibility changes, see GitHub Releases.
The current docs site history starts at 0.2.2.
0.2.6¶
Updated Installing MATE and the repository README with the 0.2.6 compatibility baseline and package-install guidance.
Added MUTLASS S4FP8 masked GroupGEMM coverage to GEMM.
Expanded the FlashAttention wrapper and DeepGEMM wrapper, including related GEMM and Python APIs coverage.
Added the FlashInfer wrapper and FlashInfer API reference for GEMM, FP8 MLA RoPE quantization, and sparse decode, together with the native Sparse MLA APIs and an end-to-end FP8 example.
Added the Development Environment page and TileLang 0.1.12 installation guidance.
0.2.5¶
Added Kimi Delta Attention (KDA) decode coverage to Python APIs and KDA.
Expanded Attention for FMHA
only_qvand MLA head-ratio coverage.Added MSA direct APIs and the MSA/fmha_sm100 wrapper for MiniMax Sparse Attention (MSA) workflows on MUSA.
Updated Installing MATE for the MUSA simple package index.
Added MUBIN artifact management commands to Command Line Interface.
Refined FlashMLA/DenseMLA support for multiple head-ratio configurations.
0.2.4¶
Updated Installing MATE for wrapper-first
pipinstallation, including automaticmatedependency installation.Added W4A8 mixed-dtype MoE GEMM to Python APIs and GEMM.
Added Mega MoE to Python APIs and Mega MoE.
0.2.3¶
Expanded the docs set around Overview, Installing MATE, Wrappers, Diagnostic Overview, Command Line Interface, Design and Architecture, and Python APIs.
Added wrapper pages for FlashAttention, SageAttention, FlashMLA, FlashKDA, and DeepGEMM.
Added API pages for Attention, GEMM, HyperConnection, and KDA.
Moved Command Line Interface, Environment Variables, and GDN Support Matrix into the Sphinx site structure.
0.2.2¶
Established the initial public docs baseline in the home page and the early API set.
Added the HyperConnection API page.
Updated FlashAttention and GDN Support Matrix.
Refreshed the home page structure.