From 505556a214e2b8580c47545439b47a8bf801b1c6 Mon Sep 17 00:00:00 2001
From: WLS2002 <64534280+WLS2002@users.noreply.github.com>
Date: Wed, 20 Aug 2025 20:26:47 +0800
Subject: [PATCH] Update README.md

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

diff --git a/README.md b/README.md
index fd3270c..95eb66e 100644
--- a/README.md
+++ b/README.md
@@ -296,6 +296,15 @@ pip install git+https://github.com/EMI-Group/tensorneat.git
 ## Multi-device and Distributed Acceleration
 TensorNEAT doesn't natively support multi-device or distributed execution, but these features can be accessed via the EvoX framework. EvoX is a high-performance, distributed, GPU-accelerated framework for Evolutionary Algorithms. For more details, visit: [EvoX GitHub](https://github.com/EMI-Group/evox/).
 
+**Notice**: As the latest EvoX has been migrated to the PyTorch backend, we need to install the JAX-Version EvoX to run multi-device EvoX. 
+The current JAX-Version Evox branch is [v0.9.1-dev](https://github.com/EMI-Group/evox/tree/v0.9.1-dev).
+
+Use 
+```bash
+pip install git+https://github.com/EMI-Group/evox/tree/v0.9.1-dev
+```
+to install the JAX based EvoX.
+
 TensorNEAT includes an EvoX Adaptor, which allows TensorNEAT algorithms to run within the EvoX framework. Additionally, TensorNEAT provides a monitor for use with EvoX.
 
 Here is an example of creating an EvoX algorithm and monitor: