# \[Blog Post\] Practical Auto-Vectorization (not LLM generated, per usual)

**URL:** <https://forums.swift.org/t/blog-post-practical-auto-vectorization-not-llm-generated-per-usual/90568>\
**Category:** Related Projects\
**Tags:** foundation\
**Created:** [November 8, 2026, 11:51am UTC](https://forums.swift.org/t/blog-post-practical-auto-vectorization-not-llm-generated-per-usual/90568 "2026-11-08T11:51:30Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![MahdiBM](https://sea2.discourse-cdn.com/swift/user_avatar/forums.swift.org/mahdibm/32/45271_2.png) [@MahdiBM](https://forums.swift.org/u/MahdiBM)\
**Post date:** [November 8, 2026, 11:51am UTC](https://forums.swift.org/t/blog-post-practical-auto-vectorization-not-llm-generated-per-usual/90568/1 "2026-11-08T11:51:30Z")

</div>

People didn't mind [[Blog Post] A Very Fast IPv4 Parser, in Swift (not LLM generated btw)](https://forums.swift.org/t/blog-post-a-very-fast-ipv4-parser-in-swift-not-llm-generated-btw/90152) to say the least, so here's another post:

> **[Practical Auto-Vectorization](https://blog.mahdibm.com/posts/practical-auto-vectorization/)**
>
> How to make your single-threaded loops run multiple times faster, or how I improved Swift's Base64 encoding performance to 16 GB/s.

This ones a bit longer than the last one, but should still be interesting.

Initially it was supposed to be a post dedicated to tips and tricks about Auto-Vectorization, but to ensure it doesn't become way too long I had to cut some parts out and let them be for future posts.

---

<div class="post-metadata">

**Author:** ![Joseph\_Heck](https://sea2.discourse-cdn.com/swift/user_avatar/forums.swift.org/joseph_heck/32/22369_2.png) [@Joseph\_Heck](https://forums.swift.org/u/Joseph_Heck)\
**Post date:** [November 8, 2026, 8:38pm UTC](https://forums.swift.org/t/blog-post-practical-auto-vectorization-not-llm-generated-per-usual/90568/2 "2026-11-08T20:38:35Z")

</div>

really great walk-through @MahdiBM - appreciate you sharing it!