前言
平日使用自己本文教大家如何部署 text-generation-webui 开源项目，简单几步让你的本地大模型拥有舒适的交互界面
conda 创建虚拟环境
创建虚拟环境，python 需要使用 3.11 的版本
conda create -n tg-webui python=3.11
进入虚拟环境
conda activate tg-webui
在 pytorch 官网找到适合自己 cuda 版本的 pytorch 安装命令，前提是要配置好适合自己显卡的 CUDA 环境，我这里 CUDA 是 11.8 ，已经提前安装好了，如果还不会可以参考我的安装教程，我使用的命令如下：
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
不同的操作系统的安装命令可以参考下面的图表内容，也可以直接去官网
clone 项目安装 requirements
从官网克隆整个项目
git clone https://github.com/oobabooga/text-generation-webui
然后进入这个文件夹中安装 requirements.txt ，各个操作系统的安装文件不一样，请自行选择。
我这里是 windows 操作系统，Nvidia 显卡，所以使用下面的安装命令。
pip install -r requirements.txt
这里需要注意 requirements.txt 最后一些无关的包（如下所示）删除即可，否则会引发不必要的错误
# llama-cpp-python (CPU only, AVX2)
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.2.81+cpuavx2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.2.81+cpuavx2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.2.81+cpuavx2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/cpu/llama_cpp_python-0.2.81+cpuavx2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
# llama-cpp-python (CUDA, no tensor cores)
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.81+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.81+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.81+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda-0.2.81+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
# llama-cpp-python (CUDA, tensor cores)
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.2.81+cu121-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.2.81+cu121-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.2.81+cu121-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/llama-cpp-python-cuBLAS-wheels/releases/download/textgen-webui/llama_cpp_python_cuda_tensorcores-0.2.81+cu121-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
# CUDA wheels
https://github.com/oobabooga/exllamav2/releases/download/v0.1.6/exllamav2-0.1.6+cu121.torch2.2.2-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.6/exllamav2-0.1.6+cu121.torch2.2.2-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.6/exllamav2-0.1.6+cu121.torch2.2.2-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.6/exllamav2-0.1.6+cu121.torch2.2.2-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
https://github.com/oobabooga/exllamav2/releases/download/v0.1.6/exllamav2-0.1.6-py3-none-any.whl; platform_system == "Linux" and platform_machine != "x86_64"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp311-cp311-win_amd64.whl; platform_system == "Windows" and python_version == "3.11"
https://github.com/oobabooga/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2.2cxx11abiFALSE-cp310-cp310-win_amd64.whl; platform_system == "Windows" and python_version == "3.10"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp311-cp311-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.11"
https://github.com/Dao-AILab/flash-attention/releases/download/v2.5.9.post1/flash_attn-2.5.9.post1+cu122torch2.2cxx11abiFALSE-cp310-cp310-linux_x86_64.whl; platform_system == "Linux" and platform_machine == "x86_64" and python_version == "3.10"
autoawq==0.2.5; platform_system == "Linux" or platform_system == "Windows"
看到下面的日志内容说明成功安装。
启动服务
控制台运行 server.py 启动服务
python .\server.py
当你看到下面的日志打印，就说明启动成功了，访问给出来的 url ： http://127.0.0.1:7860 即刻开始使用。
我们这里虽然可以启动成功，但是还没有配置可用的模型，我们将下载好的完整的 Qwen2-7B 模型，放到 text-generation-webui\models 目录下面，然后切换到 Model 页面，在 Model 框中会显示出来你放的各种模型名字，然后点击 Load 等待加载即可，加载完成即可切换到其他页面进行愉快的玩耍了。
如果在此步遇到如下问题 DLL load failed while importing flash_attn_2_cuda: The specified module could not be found，不要紧张，你只需要检查两个地方：
- 你本机的 CUDA 和你的 pytorch 版本是不是一致
- 上文提到的 requirements.txt 中文件末尾无用的行有没有删掉
如果完成了上述的两部检查，那么按照本教程重新快速部署一次环境，即可正常使用。
体验
下面我们就可以愉快的玩耍了。里面还有很多好玩的内容，可以参考官方的 WIKI 教程进一步定制化。下面是我改过名字等配置之后的效果。
