For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo package

attention

Multi-head attention (MHA) and multi-head latent attention (MLA) kernels.

Shared utilities live here (masks, operands, config), with platform-specific implementations under cpu/ and gpu/.

Packages

• ​cpu: CPU flash-attention implementation for multi-head attention.
• ​gpu: GPU multi-head attention (MHA), cross-attention, and multi-head latent attention (MLA) kernels. Vendor-specific implementations live under amd/ and nvidia/.

Modules

• ​indexer_score: Lightning-indexer scores read straight out of a paged compressed leaf.
• ​latent_sparse_attention: Sparse attention over a shared K=V latent held in two paged leaves.
• ​mha_mask: Mask types and the MHAMask trait for multi-head attention kernels.
• ​mha_operand: KV-cache operand descriptors and TMA tile builders for MHA GPU kernels.
• ​mha_utils: Shared configuration types and dispatch helpers for MHA GPU kernels.