SageAttention 产品

来源: githubOPEN_SOURCE开源CudaApache-2.0发布于 2024-10-03

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.