diff --git a/README.md b/README.md
index 0b8dcaa..5887181 100644
--- a/README.md
+++ b/README.md
@@ -45,8 +45,9 @@
   - [4. RoadMap](#4-roadmap)
   - [5. Contributions](#5-contributions)
   - [6. Acknowledgements](#6-acknowledgements)
-  - [7、Licence](#7licence)
-  - [8、Contact Information](#8contact-information)
+  - [7. Citation](#7-citation)
+  - [8. Licence](#8-licence)
+  - [9. Contact Information](#9-contact-information)
 
 ## 1. What is DB-GPT-Hub
 
@@ -89,9 +90,12 @@ DB-GPT-Hub currently supports the following base models:
   - [x] Qwen
   - [x] XVERSE
   - [x] ChatGLM2
+  - [x] ChatGLM3
   - [x] internlm
 
 
+
+
 The model is fine-tuned based on a quantization bit of 4 using Quantized Learning over Redundant Architecture (QLoRA). The minimum hardware requirements for this can be referred to as follows:   
 
 | Model Parameters | GPU RAM | CPU RAM | DISK   |
@@ -181,6 +185,7 @@ In the script, during fine-tuning, different models correspond to key parameters
 | [Qwen](https://github.com/QwenLM/Qwen-7B)                | c_attn          | chatml    |
 | [XVERSE](https://github.com/xverse-ai/XVERSE-13B)        | q_proj,v_proj   | xverse    |
 | [ChatGLM2](https://github.com/THUDM/ChatGLM2-6B)         | query_key_value | chatglm2  |
+| [ChatGLM3](https://github.com/THUDM/ChatGLM3-6B)         | query_key_value | chatglm3  |
 | [LLaMA](https://github.com/facebookresearch/llama)       | q_proj,v_proj   | -         |
 | [BLOOM](https://huggingface.co/bigscience/bloom)         | query_key_value | -         |
 | [BLOOMZ](https://huggingface.co/bigscience/bloomz)       | query_key_value | -         |
@@ -210,7 +215,7 @@ poetry run sh ./dbgpt_hub/scripts/predict_sft.sh
 ```
 
 In the script, by default with the parameter `--quantization_bit`, it predicts using QLoRA. Removing it switches to the LoRA prediction method.
-The value of the parameter  `--predicted_out_filename` is the file name of the model's predicted results, which can be found in the   `dbgpt_hub/output/pred` directory.
+The value of the parameter `predicted_input_filename`  is your predict test dataset file.  `--predicted_out_filename` is the file name of the model's predicted results.
 
 ### 3.5 Model Weights
 You can find the second corresponding model weights  from Huggingface [hg-eosphoros-ai
@@ -250,6 +255,7 @@ The whole process we will divide into three phases:
   - [x] Qwen
   - [x] XVERSE
   - [x] ChatGLM2
+  - [x] ChatGLM3
   - [x] internlm
 
 * Stage 2:
diff --git a/README.zh.md b/README.zh.md
index f29666f..4b49606 100644
--- a/README.zh.md
+++ b/README.zh.md
@@ -44,8 +44,9 @@
   - [四、发展路线](#四发展路线)
   - [五、贡献](#五贡献)
   - [六、感谢](#六感谢)
-  - [七、Licence](#七licence)
-  - [八、Contact Information](#八contact-information)
+  - [七、引用](#七引用)
+  - [八、Licence](#八licence)
+  - [九、我们的联系方式](#九我们的联系方式)
 
 ## 一、简介
 
@@ -84,6 +85,7 @@ DB-GPT-HUB目前已经支持的base模型有：
   - [x] Qwen
   - [x] XVERSE
   - [x] ChatGLM2
+  - [x] ChatGLM3
   - [x] internlm
   - [x] Falcon
 
@@ -200,7 +202,7 @@ deepspeed --num_gpus 2  dbgpt_hub/train/sft_train.py \
 poetry run sh ./dbgpt_hub/scripts/predict_sft.sh
 ```   
 脚本中默认带着参数`--quantization_bit `为QLoRA的预测，去掉即为LoRA的预测方式。  
-其中参数 `--predicted_out_filename` 的值为模型预测的结果文件名，结果在`dbgpt_hub/output/pred`目录下可以找到。
+其中参数`predicted_input_filename`  为要预测的数据集文件， `--predicted_out_filename` 的值为模型预测的结果文件名。默认结果保存在`dbgpt_hub/output/pred`目录。
 
 
 ### 3.5、模型权重
@@ -237,6 +239,7 @@ poetry run python dbgpt_hub/eval/evaluation.py --plug_value --input  Your_model_
   - [x] Qwen
   - [x] XVERSE
   - [x] ChatGLM2
+  - [x] ChatGLM3
   - [x] internlm    
   
 * 阶段二: