diff --git a/README.md b/README.md
index 6599497..66158ad 100644
--- a/README.md
+++ b/README.md
@@ -99,6 +99,9 @@ git clone https://github.com/Raykosan/ComfyUI_RaykoStudio.git
- v0.36.4 - Added drag and drop image loading to the RS Image-Text node
- v0.37.0 - Added an RS Upscaler node
- v0.38.0 - Added an RS LoRA Tester node
+- v0.39.0 - Added an RS Save Image LoRA node
+- v0.40.0 - Added RS Models Loader Pro node
+- v0.41.0 - Added RS Ref Encode node
@@ -555,6 +558,38 @@ Tested with ComfyUI core upscaler models including:
- `RealESRGAN_x4plus.pth`
- And on other models
+
+
+ 🦊 RS Ref Encode
+
+# 🦊 RS RS Ref Encode
+**A node that combines VAE Encode and Set Reference Latent into a single efficient operation.**
+
+
+
+
+
+### 🔥 Features
+- **Unified Workflow** - Replaces the 3-node chain (`VAE Encode` → `Set Reference Latent` ×2) with a single node
+- **Memory Efficient** - Encodes pixels only once and reuses the latent for both positive and negative conditioning
+- **Native Compatibility** - Uses ComfyUI's internal `node_helpers.conditioning_set_values` to ensure 100% compatibility with edit models and reference-based generation
+- **Safe VAE Handling** - Automatically handles different return formats from `vae.encode()` across various ComfyUI builds
+- **Clean API** - No extra widgets or presets. Just inputs and outputs
+
+### 🔌 Inputs & Outputs
+
+| Input | Type | Description |
+| :--- | :--- | :--- |
+| `positive` | CONDITIONING | Positive prompt conditioning |
+| `negative` | CONDITIONING | Negative prompt conditioning |
+| `pixels` | IMAGE | Reference image to encode |
+| `vae` | VAE | VAE model for encoding |
+
+| Output | Type | Description |
+| :--- | :--- | :--- |
+| `positive` | CONDITIONING | Conditioning with applied reference latent |
+| `negative` | CONDITIONING | Conditioning with applied reference latent |
+
🦊 RS Outpaint