All projects
2 repositories · updated 2026-11-07
Loading…
1thu-ml / SageAttention[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.CudaApache-2.0 · PermissiveInactiveInsufficient rank history3,999
2thu-ml / TurboDiffusionTurboDiffusion: 100–200× Acceleration for Video Diffusion ModelsPythonApache-2.0 · PermissiveInsufficient rank history3,859