SageAttention3: Microscaling FP4 Attention for Inference and An Exploration of 8-Bit Training

Documenti analoghi

• SageAttention2++: A More Efficient Implementation of SageAttention2
  di: Zhang, Jintao, et al.
  Pubblicazione: (2025)
• SageAttention: Accurate 8-Bit Attention for Plug-and-play Inference Acceleration
  di: Zhang, Jintao, et al.
  Pubblicazione: (2024)
• SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 Quantization
  di: Zhang, Jintao, et al.
  Pubblicazione: (2024)
• DGEMM without FP64 Arithmetic - Using FP64 Emulation and FP8 Tensor Cores with Ozaki Scheme
  di: Mukunoki, Daichi
  Pubblicazione: (2025)
• Gem5-AcceSys: Enabling System-Level Exploration of Standard Interconnects for Novel Accelerators
  di: Liu, Qunyou, et al.
  Pubblicazione: (2025)