doc: update requirements

README.md

@@ -124,6 +124,12 @@ The release of InternLM2 series contains two model sizes: 7B and 20B. 7B models
 
 - According to the released performance of 2024-01-17.
 
+## Requirements
+
+- Python >= 3.8
+- PyTorch >= 1.12.0 (2.0.0 and above are recommended)
+- Transformers >= 4.34
+
 ## Usages
 
 We briefly show the usages with [Transformers](#import-from-transformers), [ModelScope](#import-from-modelscope), and [Web demos](#dialogue).
@@ -143,7 +149,7 @@ tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_re
 # Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
 model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
 # (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
-  # InternLM 7B in 4bit will cost nearly 8GB GPU memory. 
+  # InternLM 7B in 4bit will cost nearly 8GB GPU memory.
   # pip install -U bitsandbytes
   # 8-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_8bit=True)
   # 4-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_4bit=True)
@@ -167,7 +173,7 @@ tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_re
 # Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
 model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
 # (Optional) If on low resource devices, you can load model in 4-bit or 8-bit to further save GPU memory via bitsandbytes.
-  # InternLM 7B in 4bit will cost nearly 8GB GPU memory. 
+  # InternLM 7B in 4bit will cost nearly 8GB GPU memory.
   # pip install -U bitsandbytes
   # 8-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_8bit=True)
   # 4-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_4bit=True)
@@ -183,7 +189,7 @@ print(response)
 You can interact with the InternLM Chat 7B model through a frontend interface by running the following code:
 
 ```bash
-pip install streamlit==1.24.0
+pip install streamlit
 pip install transformers>=4.34
 streamlit run ./chat/web_demo.py
 ```
@@ -192,7 +198,7 @@ streamlit run ./chat/web_demo.py
 
 We use [LMDeploy](https://github.com/InternLM/LMDeploy) for fast deployment of InternLM.
 
-With only 4 lines of codes, you can perform `internlm2-chat-7b` inference after `pip install lmdeploy`.
+With only 4 lines of codes, you can perform `internlm2-chat-7b` inference after `pip install lmdeploy>=0.2.1`.
 
 ```python
 from lmdeploy import pipeline

README_zh-CN.md

@@ -122,6 +122,12 @@ InternLM2 系列模型在本仓库正式发布，具有如下特性：
 
 - 性能数据截止2024-01-17
 
+## 依赖
+
+- Python >= 3.8
+- PyTorch >= 1.12.0 (推荐 2.0.0 和更高版本)
+- Transformers >= 4.34
+
 ## 使用案例
 
 接下来我们展示使用 [Transformers](#import-from-transformers)，[ModelScope](#import-from-modelscope) 和 [Web demo](#dialogue) 进行推理。
@@ -141,7 +147,7 @@ tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-chat-7b", trust_re
 # 设置`torch_dtype=torch.float16`来将模型精度指定为torch.float16，否则可能会因为您的硬件原因造成显存不足的问题。
 model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-chat-7b", device_map="auto",trust_remote_code=True, torch_dtype=torch.float16)
 # (可选) 如果在低资源设备上，可以通过bitsandbytes加载4-bit或8-bit量化的模型，进一步节省GPU显存.
-  # 4-bit 量化的 InternLM 7B 大约会消耗 8GB 显存. 
+  # 4-bit 量化的 InternLM 7B 大约会消耗 8GB 显存.
   # pip install -U bitsandbytes
   # 8-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_8bit=True)
   # 4-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_4bit=True)
@@ -164,7 +170,7 @@ model_dir = snapshot_download('Shanghai_AI_Laboratory/internlm2-chat-7b')
 tokenizer = AutoTokenizer.from_pretrained(model_dir, device_map="auto", trust_remote_code=True)
 model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, torch_dtype=torch.float16)
 # (可选) 如果在低资源设备上，可以通过bitsandbytes加载4-bit或8-bit量化的模型，进一步节省GPU显存.
-  # 4-bit 量化的 InternLM 7B 大约会消耗 8GB 显存. 
+  # 4-bit 量化的 InternLM 7B 大约会消耗 8GB 显存.
   # pip install -U bitsandbytes
   # 8-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_8bit=True)
   # 4-bit: model = AutoModelForCausalLM.from_pretrained(model_dir, device_map="auto", trust_remote_code=True, load_in_4bit=True)
@@ -180,7 +186,7 @@ print(response)
 可以通过以下代码启动一个前端的界面来与 InternLM Chat 7B 模型进行交互
 
 ```bash
-pip install streamlit==1.24.0
+pip install streamlit
 pip install transformers>=4.34
 streamlit run ./chat/web_demo.py
 ```
@@ -189,7 +195,7 @@ streamlit run ./chat/web_demo.py
 
 我们使用 [LMDeploy](https://github.com/InternLM/LMDeploy) 完成 InternLM 的一键部署。
 
-通过 `pip install lmdeploy` 安装 LMDeploy 之后，只需 4 行代码，就可以实现离线批处理：
+通过 `pip install lmdeploy>=0.2.1` 安装 LMDeploy 之后，只需 4 行代码，就可以实现离线批处理：
 
 ```python
 from lmdeploy import pipeline

chat/README.md

@@ -51,8 +51,8 @@ print(response)
 You can interact with the InternLM Chat 7B model through a frontend interface by running the following code:
 
 ```bash
-pip install streamlit==1.24.0
-pip install transformers==4.30.2
+pip install streamlit
+pip install transformers>=4.34
 streamlit run ./chat/web_demo.py
 ```
 

chat/README_zh-CN.md

@@ -45,7 +45,7 @@ print(response)
 可以通过以下代码启动一个前端的界面来与 InternLM2 Chat 7B 模型进行交互
 
 ```bash
-pip install streamlit==1.24.0
-pip install transformers==4.30.2
+pip install streamlit
+pip install transformers>=4.34
 streamlit run ./web_demo.py
 ```

chat/lmdeploy.md

@@ -12,7 +12,7 @@ This article primarily highlights the basic usage of LMDeploy. For a comprehensi
 Install lmdeploy with pip (python 3.8+)
 
 ```shell
-pip install lmdeploy
+pip install lmdeploy>=0.2.1
 ```
 
 ## Offline batch inference

chat/lmdeploy_zh_cn.md

@@ -12,7 +12,7 @@
 使用 pip（python 3.8+）安装 LMDeploy
 
 ```shell
-pip install lmdeploy
+pip install lmdeploy>=0.2.1
 ```
 
 ## 离线批处理

finetune/README.md

@@ -29,7 +29,7 @@ We recommend two projects to fine-tune InternLM.
 - Install XTuner with DeepSpeed integration
 
   ```shell
-  pip install -U 'xtuner[deepspeed]'
+  pip install -U 'xtuner[deepspeed]>=0.1.13'
   ```
 
 ### Fine-tune

finetune/README_zh-CN.md

@@ -29,7 +29,7 @@
 - 安装集成 DeepSpeed 版本的 XTuner
 
   ```shell
-  pip install -U 'xtuner[deepspeed]'
+  pip install -U 'xtuner[deepspeed]>=0.1.13'
   ```
 
 ### 微调