diff --git a/docs/dockerfile/dockerfile b/docs/dockerfile/dockerfile
new file mode 100644
index 0000000000000000000000000000000000000000..bfc991efb22a07f88af1ef2910d562e4450468dd
--- /dev/null
+++ b/docs/dockerfile/dockerfile
@@ -0,0 +1,73 @@
+FROM ubuntu:22.04 as rl_base
+
+ARG DEBIAN_FRONTEND=noninteractive
+ENV TZ=Asia/Shanghai
+
+WORKDIR /root
+
+RUN echo 'Acquire::AllowInsecureRepositories “true”; ' >> /etc/apt/apt.conf.d/allow-insecure && \
+    echo 'Acquire::AllowDowngradeToInsecureRepositories “true”; ' >> /etc/apt/apt.conf.d/allow-insecure
+
+RUN apt-get update --allow-unauthenticated --allow-insecure-repositories && \
+    apt-get install -y libsox-dev unzip libaio-dev zip iputils-ping telnet sudo git wget net-tools python3.10 python3-pip libjemalloc2 --allow-unauthenticated  && \
+    apt-get clean && \
+    rm -rf /var/lib/apt/lists/*
+
+# 此处可以选择配置pip镜像源
+RUN pip3 install --upgrade pip
+RUN pip3 install cffi pathlib2 absl-py
+
+# 创建软连接
+RUN ln -s /usr/bin/python3.10 /usr/bin/python && \
+    rm /usr/bin/pip && \
+    ln -s /usr/bin/pip3.10 /usr/bin/pip
+
+RUN echo "\n" >> ~/.bashrc  && \
+    echo "export LD_PRELOAD=/usr/lib/aarch64-linux-gnu/libjemalloc.so.2:\${LD_PRELOAD}" >> ~/.bashrc && \
+    git config --global http.sslVerify false
+
+# 添加非root用户
+RUN umask 0022 && groupadd HwHiAiUser && useradd -g HwHiAiUser -d /home/HwHiAiUser -m HwHiAiUser -s /bin/bash
+
+RUN wget https://pytorch-package.obs.cn-north-4.myhuaweicloud.com/pta/community-edition/20250630-0630poc/pytorch_v2.6.0.zip -O torch.zip --no-check-certificate && \
+    unzip torch.zip && \
+    tar -xzvf pytorch_v2.6.0_py310.tar.gz && \
+    pip install torch_npu-2.6.0.dev20250612-cp310-cp310-manylinux_2_28_aarch64.whl
+
+RUN pip uninstall setuptools -y \
+    pip install setuptools==69.5.1 && \
+    git clone https://github.com/vllm-project/vllm.git  && \
+    cd vllm && \
+    git checkout 5bc1ad6cee754405464a9957e86cf3a9302e4986 && \
+    VLLM_TARGET_DEVICE=empty pip install -e . && \
+    cd .. && \
+    git clone -b main https://github.com/vllm-project/vllm-ascend.git && \
+    cd vllm-ascend && \
+    git checkout 908a851 && \
+    COMPILE_CUSTOM_KERNELS=0 pip install -e .
+
+RUN git clone https://gitee.com/ascend/MindSpeed-RL.git && \
+    cd  MindSpeed-RL && \
+    git checkout acb2c70182283  && \
+    cd ../  && \
+    git clone https://gitee.com/ascend/MindSpeed.git && \
+    cd  MindSpeed && \
+    git checkout 2c085cc9  && \
+    pip install -r requirements.txt && \
+    cp -r mindspeed ../MindSpeed-RL/&& \
+    cd .. && \
+    git clone https://github.com/NVIDIA/Megatron-LM.git  && \
+    cd Megatron-LM && \
+    git checkout core_r0.8.0 && \
+    cp -r megatron ../MindSpeed-RL/ && \
+    cd .. && \
+    git clone https://gitee.com/ascend/MindSpeed-LLM.git && \
+    cd MindSpeed-LLM && \
+    git checkout 1cd73e869f4c && \
+    cp -r mindspeed_llm ../MindSpeed-RL/ && \
+    cd .. && \
+    cd ./MindSpeed-RL && \
+    pip install -r requirements.txt && \
+    pip install antlr4-python3-runtime==4.7.2 --no-deps && \
+    cd ../ && \
+    pip install torch_npu-2.6.0.dev20250612-cp310-cp310-manylinux_2_28_aarch64.whl
diff --git a/docs/dockerfile/dockerfile.md b/docs/dockerfile/dockerfile.md
new file mode 100644
index 0000000000000000000000000000000000000000..83e619d5c45c53544857254c3a3418430dbb28c2
--- /dev/null
+++ b/docs/dockerfile/dockerfile.md
@@ -0,0 +1,41 @@
+# 镜像构建
+
+在包含dockerfile的目录下运行：
+
+```shell
+docker build -t your_image_name:tag .
+```
+
+# 创建镜像容器
+
+在使用此dockerfile之前，请安装配套的昇腾[CANN](https://support.huawei.com/carrier/navi?coltype=software#col=software&detailId=PBI1-265642021&path=PBI1-262732867/PBI1-262735886/PBI1-22892969/PBI1-23710427/PBI1-251168373)。
+CANN的安装请参考[CANN安装](../install_guide.md)：
+
+```shell
+# 挂载镜像
+docker run -dit --name 'rl_test' -v /usr/local/Ascend/driver:/usr/local/Ascend/driver -v /usr/local/Ascend/ascend:/usr/local/Ascend/ascend -v /usr/local/Ascend/firmware:/usr/local/Ascend/firmware  -v /usr/local/sbin/:/usr/local/sbin/ -v /home/:/home/ your_image_name:tag bash
+```
+
+# 登录镜像并确认环境状态
+
+```shell
+# 登录容器
+docker exec -it rl_test /bin/bash                           
+# 确认npu是否可以正常使用，否则返回3.检查配置
+npu-smi info
+```
+
+# 编译以及安装apex
+
+```shell
+git clone -b master https://gitee.com/ascend/apex.git
+cd apex/
+bash scripts/build.sh --python=3.10
+
+cd dist/
+pip uninstall apex
+pip install --upgrade apex-0.1+ascend-{version}.whl # version为python版本和cpu架构
+```
+
+# 单机以及多机模型的预训练任务运行
+基于镜像和仓库代码，完成环境部署，可执行单机和多机的预训练任务。