/SageAttention

Quantized Attention that achieves speedups of 2.1x and 2.7x compared to FlashAttention2 and xformers, respectively, without lossing end-to-end metrics across various models.

Primary LanguagePythonBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

Stargazers

No one’s star this repository yet.