From eefaa5b0580483425751cfa8d342716c8f751d16 Mon Sep 17 00:00:00 2001
From: obisin <90648621+obisin@users.noreply.github.com>
Date: Wed, 20 Aug 2025 15:30:14 +0100
Subject: [PATCH] Update README.md

---
 README.md | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index ab9b1d3..67115eb 100644
--- a/README.md
+++ b/README.md
@@ -175,12 +175,13 @@ NOTE: All settings can be stacked together!
 - **Purpose**: Background thread for automatic layer management
 - **Benefit**: Better overlap of transfers with computation
 - **Note**: If you're using a smaller model or one with a fast compute per layer, threading may be counter productive as the swap will lag behind the fast compute. Instead increase resident GPU layers and prefetch
-- **Note**: May not work with all models, Flux can sometimes have issues. May cause instability in some systems.
+- **Note**: May cause instability in some systems
 
 **`cuda_streams`**
 - **Purpose**: CUDA streams for copy-compute overlap
 - **Requirement**: Additional VRAM for stream buffers
 - **Benefit**: Maximum performance when VRAM allows
+- **Note**: May not work with all models, Flux can sometimes have issues
 
 **`batch_move`**
 - **Purpose**: Move multiple layers simultaneously