diff --git a/README.md b/README.md
index aff5067..b2ce05c 100644
--- a/README.md
+++ b/README.md
@@ -56,7 +56,7 @@ Anima模型基于QLoRA开源的[33B guanaco](https://huggingface.co/timdettmers/
 
 #### 如何训练
 
-使用以下步骤可以重现Anima 33B模型：
+使用以下步骤可以重现Anima 33B模型（单卡80GB H100或双卡 40GB A100均测试过可运行）：
 
 ```bash
 # 1. install dependencies
@@ -66,6 +66,9 @@ cd training
 ./run_Amina_training.sh
 ```
 
+#### 多卡训练
+由于使用Hugging Face Accelerate，天然支持多卡训练。
+我们测试过双卡40GB的A100，可以直接运行。
 
 ## 📊验证评估
 
diff --git a/README_en.md b/README_en.md
index df7c9f8..05f4b6e 100644
--- a/README_en.md
+++ b/README_en.md
@@ -57,7 +57,7 @@ For cost considerations, we mostly chose not to do too much grid search, assumin
 
 #### How to reproduce our training
 
-Anima 33B model could be reproduced fully with the following steps：
+Anima 33B model could be reproduced fully with the following steps(tested on single GPU environment of 1x80GB H100, or multi-GPU of 2xA100 40GB)：
 
 ```bash
 # 1. install dependencies
@@ -66,7 +66,10 @@ pip install -r requirements.txt
 cd training
 ./run_Amina_training.sh
 ```
+#### Multi-GPU training
+Bause of Hugging Face Accelerate，multi-GPU training is supported out-of-box.
 
+We tested 2xA100 40GB, the above script can work directlly seemlessly.
 
 ## 📊Evaluations