LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Other

flash-attention

Fast and memory-efficient exact attention

OtherEmerging

Stars

—

Forks

—

Contributors

8

Last push

2mo ago

Recent commits

Latest commits.

SM90 FA4 QuACK 0.4 Compatibility (#2513)
ebeff90EduardDurech2mo ago
[hd256] Improve forward kernel with exp2 FMA emulation (+3% to +9% performance gain) (#2488)
6c73fb5Johnsonms2mo ago
Add cache utils logging test (#2509)
96bd151Driss Guessous2mo ago
[Tests,MLA] Close coverage gaps in test_flash_attn_mla_absorbed{,_varlen} (#2483)
89ce84bJohnsonms2mo ago
Fix clc scheduling request bug (#2508)
b86e0ccDriss Guessous2mo ago

519445aMatthew Bonanni2mo ago

[CuTe, Flex] simplify blocksparse interface in flash_attn_func (#2506)

6b52632Reuben Stern2mo ago

[SM100] Guard gO None in empty-tile correction (#2504)

547031ageruome2mo ago

Top contributors

Builders behind this project.

guilhermeleobas