Skip to main content
Search

# Online Softmax

1 related articles

Subscribe to updates
  1. arXiv

    Atik puts matmul and attention on one fixed-point MAC mesh, reaching 225.8x peak attention speedup on a Rocket core

    The open-source accelerator Atik attaches to a RISC-V RocketChip core over RoCC and runs dense matrix multiplication, attention scoring and probability-value accumulation on one shared fixed-point MAC mesh, while tensors enter and leave as BF16. Against the BF16 scalar reference on the same Rocket core, the 88 configuration reaches 31.5x peak matmul speedup and 225.8x peak attention speedup in cycle-accurate FireSim runs; the 44.2, 77.1 and 32.3 figures for ViT, TinyBERT and GPT-2 prefill are hybrid projections, not full-model hardware replays.

You’re all caught up