Zyphra ZYFA FP8 FlashAttention for AMD MI355X up to 1.76x faster
ZYFA achieves up to 1.76x faster performance using FP8 FlashAttention on AMD MI355X.
Stories per day · 30 days
3 RSS
Relatedamd1attention1coding-agents1cuda1fp81gpu-kernels1kernel-optimization1kernels1long-context1mi355x1relative-position-attention1sql1transformers1zyphra1
8 stories, newest first · one card per story
Stories tagged #flashattention
KernelSeek beats FlashAttention-3 by asking questions, not adding tools
A tool-rich agent stalled at 0.50x after 133 steps; KernelSeek reached 1.13x, then 1.16x and 1.46x on later kernels.
Show HN derives a FlashAttention kernel from about 25 lines of SQL
The Hacker News feed only shows the title, 20 points, and one comment.
Fast Relative-Position Attention Algorithm Speeds Up Transformers
Exact tiled computation incorporates additive position terms into FlashAttention without redundant indexing or memory overhead.
FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention
FlashMask-3 adds Hopper and distributed mask-aware attention, raising long-context training throughput over prior kernels.
Approximating Softmax in Pretrained LLMs: Model Sensitivity and Kernel Acceleration
Approximating softmax in ten frozen LLMs yields a FlashAttention-4 kernel up to 25.8% faster on NVIDIA B200.
SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking
SAS trains a gated sparse attention selector end-to-end with the language modeling loss, beating distillation-based sparsifiers on reasoning, long-context, and agentic tasks.
Perplexity Details Its GPU Embedding Stack: How Ivy, Tulip and ROSE Serve pplx-embed
Perplexity details its GPU embedding serving stack (Ivy, Tulip, ROSE), which reuses LLM prefill/decode kernels, CUDA graphs, and LazyTensors to cut launch overhead.
4