Understanding FlashAttention Pt 1: Personal Notes
Introduction How IO-Aware Attention Makes Transformers Faster Without Approximating Attention The mechanism, in three words: Tiling + Online Softmax + Recomputation. Everything in this handbook is elaboration on that summary. A technical handbook on exact tiled attention: GPU memory traffic, online softmax, forward and backward passes, IO complexity, the evolution from FlashAttention-1 through FlashAttention-4, and current framework behavior.
- ▪Introduction How IO-Aware Attention Makes Transformers Faster Without Approximating Attention The mechanism, in three words: Tiling + Online Softmax + Recomputation.
- ▪Everything in this handbook is elaboration on that summary.
- ▪A technical handbook on exact tiled attention: GPU memory traffic, online softmax, forward and backward passes, IO complexity, the evolution from FlashAttention-1 through FlashAttention-4, and current framework behavior.
Hacker News (Front Page) files mainly under programming. We currently carry 1,611 of its stories. Top-voted stories on Hacker News.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | Github |
| Canonical URL | https://chizkidd.github.io//2026/09/13/flashattention/ |
| Publication time | Mon, 14 Sep 2026 14:01:04 +0000 |
| Retrieval time | 2026-09-14T14:56:51.338Z |
| Last seen | 2026-09-14T14:56:51.338Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | rm-Cu7c6DJuX · 1 stories |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
0. Introduction How IO-Aware Attention Makes Transformers Faster Without Approximating Attention The mechanism, in three words: Tiling + Online Softmax + Recomputation. Everything in this handbook is elaboration on that summary. A technical handbook on exact tiled attention: GPU memory traffic, online softmax, forward and backward passes, IO complexity, the evolution from FlashAttention-1 through FlashAttention-4, and current framework behavior. 0.1 How to Read This Handbook This handbook was inspired by this tweet. Before the fix, here is what the standard attention implementation looks like. Load $Q, K, V \in \mathbb{R}^{N \times d}$ in HBM, then: Read $Q, K$ from HBM, compute $S$, write $S$ to HBM. Read $S$ from HBM, compute $P$, write $P$ to HBM.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Github.
Discussion
0 commentsMore from Github
![Everything is a UOp – reading the deep learning stack behind comma[dot]AI](/img-proxy?url=https%3A%2F%2Fssenthilnathan3.github.io%2Fimages%2Ftinygrad.jpeg&w=320)
