From e4cc9b70166eeb6faaa3ecfb249a01306847073d Mon Sep 17 00:00:00 2001
From: ussoewwin <136552381+ussoewwin@users.noreply.github.com>
Date: Sat, 15 Nov 2025 15:01:49 +0900
Subject: [PATCH] v1.63: Add user-configurable CPU offload setting (Issue #21)

---
 README.md | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index b9c16fd..cd70868 100644
--- a/README.md
+++ b/README.md
@@ -324,7 +324,14 @@ This node is designed to work with:
 
 ## Changelog
 
-### v1.62 (Latest)
+### v1.63 (Latest)
+- **Fixed**: Addressed [Issue #21](https://github.com/ussoewwin/ComfyUI-QwenImageLoraLoader/issues/21) – User-configurable CPU offload setting
+- **Problem**: CPU offload setting was hardcoded to `"auto"`, causing unnecessary slowdowns when VRAM was sufficient
+- **Solution**: Added `cpu_offload` parameter to `INPUT_TYPES` allowing users to select from `["auto", "enable", "disable"]` with default `"disable"` for performance
+- **Current Status**: ✅ **Fully resolved** - Users can now control CPU offload settings through the node UI
+- **Technical Details**: See [v1.63 Release Notes](https://github.com/ussoewwin/ComfyUI-QwenImageLoraLoader/releases/tag/v1.63) for complete explanation
+
+### v1.62
 - **Attempted Fix**: Addressed [Issue #14](https://github.com/ussoewwin/ComfyUI-QwenImageLoraLoader/issues/14) – Multi-stage workflow cache not resetting when LoRAs change
 - **Problem**: Cache was not being reset when switching between different LoRA sets in multi-stage workflows, causing incorrect results
 - **Solution Attempted**: Cache invalidation logic was added to reset cache when LoRAs change