SwiftCIM: a 55nm 23.2μJ/Token L-0.5 ReRAM Coupled Digital CIM Accelerator with Fully-Fused Multi-Head Attention Dataflow for FlashAttention
Published in ESSERC, 2026
Recommended citation: Kunming Shao, Xiaomeng Wang, and collaborators. SwiftCIM: a 55nm 23.2μJ/Token L-0.5 ReRAM Coupled Digital CIM Accelerator with Fully-Fused Multi-Head Attention Dataflow for FlashAttention. In 2026 IEEE European Solid-State Electronics Research Conference (ESSERC), 2026. https://epapers2.org/esserc2026/ESR/paper_details.php?paper_id=8143
SwiftCIM: a 55nm 23.2μJ/Token L-0.5 ReRAM Coupled Digital CIM Accelerator with Fully-Fused Multi-Head Attention Dataflow for FlashAttention
