flashdecoding

Here is 1 public repository matching this topic...

ankur110 / triton-attention-lab

FlashAttention-2 forward and Flash-Decoding kernels written from scratch in Triton, with grouped-query attention (GQA) and causal masking, benchmarked against PyTorch SDPA and a naive eager implementation, then run end-to-end inside a from-scratch Llama-3.2-1B.

benchmarking flash kernel decoding speed triton autotuning profiling sdpa gqa kv-cache fa2 flash-attention flashdecoding

Updated Sep 21, 2026
Jupyter Notebook

Add this topic to your repo

To associate your repository with the flashdecoding topic, visit your repo's landing page and select "manage topics."

Learn more