API Reference

GEMM Kernels

Attention Kernels

Normalization

Convolution

Sparse Operations

Quantization

On this page

Are you an LLM? You can read better optimized documentation at /modern-ai-kernels/en/api/attention.md for this page in Markdown format

Attention Kernels

This page is under construction.

FlashAttention-style memory-efficient attention implementation.

Coming soon.

See Papers & Citations for FlashAttention papers.

Last updated:

Pager

Released under the Apache 2.0 License.