diff --git a/ch05/16_qwen3.5/README.md b/ch05/16_qwen3.5/README.md
index 0a55245..4b3b5af 100644
--- a/ch05/16_qwen3.5/README.md
+++ b/ch05/16_qwen3.5/README.md
@@ -2,7 +2,7 @@
 
 This folder contains a from-scratch style implementation of [Qwen/Qwen3.5-0.8B](https://huggingface.co/Qwen/Qwen3.5-0.8B).
 
-<img src="https://sebastianraschka.com/images/LLMs-from-scratch-images/bonus/qwen3.5/01.webp" width="500px">
+<img src="https://sebastianraschka.com/images/LLMs-from-scratch-images/bonus/qwen3.5/03.webp">
 
 Qwen3.5 is based on the Qwen3-Next architecture, which I described in more detail in section [2. (Linear) Attention Hybrids](https://magazine.sebastianraschka.com/i/177848019/2-linear-attention-hybrids) of my [Beyond Standard LLMs](https://magazine.sebastianraschka.com/p/beyond-standard-llms) article