Ollama

安装

Docker 安装

docker run -d -v /data/ollama:/root/.ollama -p 8080:11434 --name ollama ollama/ollama
docker exec -it ollama ollama run qwen2:1.5b

联网安装

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

离线安装

#### 下载对应版本
## ARM
curl -fsSL https://ollama.com/download/ollama-linux-arm64.tgz | sudo tar zx -C /usr
## AMD
curl -fsSL https://ollama.com/download/ollama-linux-amd64.tgz | sudo tar zx -C /usr
## AMD+GPU
curl -fsSL https://ollama.com/download/ollama-linux-amd64-rocm.tgz | sudo tar zx -C /usr

## 创建用户
sudo useradd -r -s /bin/false -m -d /usr/share/ollama ollama
## 创建服务文件
vim /etc/systemd/system/ollama.service
## 启动
sudo systemctl daemon-reload
sudo systemctl start ollama
sudo systemctl enable ollama

服务文件

[Unit]
Description=Ollama Service
After=network-online.target

[Service]
ExecStart=/usr/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3

[Install]
WantedBy=default.target

命令

## 下载
ollama pull llama3.1
## 删除
ollama rm llama3.1
## 复制
ollama cp llama3.1 my-model
## 已安装
ollama list
## 已启动的
ollama ps
## 详情
ollama show llama3.1

## 启动服务
ollama serve

本地模型

vim qwen1_5-0_5b.mf
ollama create qwen1_5-0_5b -f qwen1_5-0_5b.mf

FROM /models/Qwen1.5-32B-Chat-GGUF/qwen1_5-32b-chat-q4_k_m.gguf
FROM /data/llm/models/qwen1_5-0_5b-chat-q4_k_m.gguf

TEMPLATE """{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>{{ end }}<|im_start|>user
{{ .Prompt }}<|im_end|>
<|im_start|>assistant
"""
PARAMETER stop "<|im_start|>"
PARAMETER stop "<|im_end|>"

存储位置

默认位置：

• macOS: ~/.ollama/models
• Linux: /usr/share/ollama/.ollama/models
• Windows: C:\Users<username>.ollama\models

更改

vim /etc/systemd/system/ollama.service

配置

[Unit]
Description=Ollama Service
After=network-online.target

[Service]
ExecStart=/usr/bin/ollama serve
User=ollama
Group=ollama
Restart=always
RestartSec=3
## 增加 外网访问、模型目录
Environment="OLLAMA_HOST=0.0.0.0"
Environment="OLLAMA_MODELS=/app/models/ollama"

[Install]
WantedBy=default.target