Skip to content

Commit

Permalink
add big model doc
Browse files Browse the repository at this point in the history
  • Loading branch information
刘金发 committed Feb 28, 2025
1 parent d9ea0e7 commit 28d93c7
Show file tree
Hide file tree
Showing 2 changed files with 114 additions and 0 deletions.
15 changes: 15 additions & 0 deletions source/_posts/materiel/ai/ollama/doc.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# ollama使用记录

[官网](https://ollama.com)

## 安装

```bash
curl -fsSL https://ollama.com/install.sh | sh
```

## 使用

```bash
ollama run BAAI/bge-reranker-v2-m3
```
99 changes: 99 additions & 0 deletions source/_posts/tools/terminal/vllm/doc.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,99 @@
# vllm使用记录

[GitHub地址](https://github.com/vllm-project/vllm)

## 安装

```bash
# 安装 Miniconda
# 运行安装脚本:

# Linux:
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh -b -p ${HOME}/software/miniconda3
# macOS:
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-x86_64.sh
bash Miniconda3-latest-MacOSX-x86_64.sh -b -p ${HOME}/software/miniconda3

# 配置环境变量:

# 将以下内容添加到 ~/.bashrc 或 ~/.zshrc 文件中:
export PATH=${HOME}/software/miniconda3/bin:$PATH
# 刷新环境变量:
source ~/.bashrc
# 验证安装:
conda --version
# 如果显示版本号,说明安装成功。
```

```bash
# (Recommended) Create a new conda environment.
conda create -n vllm python=3.12 -y
conda activate vllm
# PS: 有时可能需要先初始化 conda init 然后执行 bash 重新打开终端, 之后才能使用 conda activate vllm
# 取消激活环境
conda deactivate


# 配置 pip 源
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple/
pip config set install.trusted-host pypi.tuna.tsinghua.edu.cn


# 在虚拟环境中安装 vllm 0.4.2 版本
pip install vllm==0.4.2
# 或者安装最新版本
pip install vllm
```

- docker 安装

```bash
# vLLM 提供了一个官方 Docker 镜像用于部署。该镜像可用于运行与 OpenAI 兼容服务器,并且可在 Docker Hub 上以 vllm/vllm-openai 的形式获取。
# PS: 需要英伟达运行时
docker run --runtime nvidia --gpus all \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HUGGING_FACE_HUB_TOKEN=<secret>" \
-p 8000:8000 \
--ipc=host \
vllm/vllm-openai:latest \
--model mistralai/Mistral-7B-v0.1
```

- 源码安装

```bash
# 启用Docker BuildKit
# export DOCKER_BUILDKIT=1
# export COMPOSE_DOCKER_CLI_BUILD=1
docker build -f Dockerfile.cpu -t vllm-cpu-env --shm-size=4g .
docker run -it \
--rm \
--network=host \
--cpuset-cpus=<cpu-id-list, optional> \
--cpuset-mems=<memory-node, optional> \
vllm-cpu-env
```

## 使用

```bash
# Run the following command to start the vLLM server with the Qwen2.5-1.5B-Instruct model:
vllm serve Qwen/Qwen2.5-1.5B-Instruct
vllm serve BAAI/bge-reranker-v2-m3
```

## 异常问题

- `ModuleNotFoundError: No module named 'torch'`

```bash
pip install torch
```

- for instructions on how to install GCC 5 or higher.

```bash
yum install centos-release-scl
yum install devtoolset-11-gcc devtoolset-11-gcc-c++
```

0 comments on commit 28d93c7

Please sign in to comment.