# llama.cpp adds flash attention operation

> The open‑source llama.cpp library now supports a flash attention op via ggml‑zdnn, announced on 20 Oct 2026.

Oossa · 2026-10-20 · https://oossa.com/en/llama-cpp-adds-flash-attention-operation

The ggml‑org team released llama.cpp version b11759 on 20 Oct 2026. The update adds a flash attention operation (fattn) to the ggml‑zdnn backend and replaces scratch slots with a scratch buffer pool. The change is signed off by contributor Aaron Teo. Flash attention speeds up the attention calculation that large language models use.

## The facts

- Flash attention op (fattn) added to ggml‑zdnn
- Release date: 2026‑10‑20

## Why it matters

Flash attention can cut inference time for users running llama.cpp locally, making chat‑style models feel faster.

## Sources & references

1. [ggml-org/llama.cpp b11759](https://github.com/ggml-org/llama.cpp/releases/tag/b11759) – llama.cpp, 2026-10-20

Last updated: 2026-10-20