wsl部署AI模型教程+虚拟环境配置教程

模型路径：/mnt/d/DEEPSEEK/models/QwQ-32B-AWQ。

安装python3：

sudo apt update
sudo apt install python3.12 python3.12-venv python3.12-dev build-essential libssl-dev libffi-dev

安装python3的赖：
sudo apt install python3.12 python3.12-venv python3.12-dev build-essential libssl-dev libffi-dev

# 创建符号链接（如果需要）
sudo ln -s /usr/bin/python3.12 /usr/bin/python3

# 验证符号链接
which python3

# 创建虚拟环境
python3 -m venv qwen_env

# 激活虚拟环境
source qwen_env/bin/activate

# 验证虚拟环境
python –version
pip –version

安装numpy：pip install numpy

手动下载安装：pip install /mnt/d/cnn/torch-2.6.0+cu118-cp312-cp312-linux_x86_64.whl
pip install /mnt/d/cnn/nvidia_cuda_cupti_cu11-11.8.87-py3-none-manylinux2014_x86_64.whl
pip install /mnt/d/cnn/nvidia_cuda_nvrtc_cu11-11.8.89-py3-none-manylinux2014_x86_64.whl
pip install /mnt/d/cnn/triton-3.2.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl

切换到模型目录：
source qwen_env/bin/activate
cd /mnt/d/DEEPSEEK/models/QwQ-32B-AWQ

source qwen_env/bin/activate
cd /home/dministrator/Qwen2.5-3B-Instruct
python api_server.py

******************************************
可视化web训练LLaMA-Factory
开源地址：https://github.com/hiyouga/LLaMA-Factory?tab=readme-ov-file
source qwen_env/bin/activate
cd /home/dministrator/LLaMA-Factory-0.9.1
pip install -e “.[torch,metrics]” -i https://pypi.tuna.tsinghua.edu.cn/simple
llamafactory-cli webui
http://localhost:7860/
——————————