From d9f3ef335a391056632d8ffc02c5b848d08a72f8 Mon Sep 17 00:00:00 2001
From: Scott Roy <161522778+metascroy@users.noreply.github.com>
Date: Wed, 23 Oct 2024 13:57:30 -0700
Subject: [PATCH] Update quantization.md

---
 docs/quantization.md | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/docs/quantization.md b/docs/quantization.md
index bef7309c6..3415d8cb8 100644
--- a/docs/quantization.md
+++ b/docs/quantization.md
@@ -120,6 +120,8 @@ python3 torchchat.py generate llama3 --pte-path llama3.pte  --prompt "Hello my n
 
 ## Experimental TorchAO lowbit kernels
 
+WARNING: These kernels only work on devices with ARM CPUs, for example on Mac computers with Apple Silicon.
+
 ### Use
 
 #### linear:a8wxdq