欣淇
发布于 2026-05-20 / 0 阅读
0
0

🚀 Ollama:171,751 stars · 一键运行开源大模型的统一平台

🚀 Ollama:171,751 stars · 一键运行开源大模型的统一平台

简介

Ollama 是一个开源的大模型运行平台,让你轻松在本地运行 Kimi-K2.5、GLM-5、MiniMax、DeepSeek、Gemma 等各种开源模型。它简化了模型的部署和使用过程,让开发者能够快速上手 AI 开发。

快速开始

安装

macOS:

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

Linux:

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

Windows:

irm https://ollama.com/install.ps1 | iex

基本使用

启动 Ollama:

ollama

运行特定模型:

ollama run gemma3

集成开发

Ollama 支持与多种 AI 工具集成:

Claude Code 集成:

ollama launch claude

OpenClaw 助手:

ollama launch openclaw

支持的集成包括 Codex、Copilot CLI、Droid、OpenCode 等。

REST API

Ollama 提供完整的 REST API:

curl http://localhost:11434/api/chat -d '{
  "model": "gemma3",
  "messages": [{
    "role": "user",
    "content": "Why is the sky blue?"
  }],
  "stream": false
}'

优势

  • 简单易用: 一行命令即可运行模型
  • 多平台支持: macOS、Windows、Linux、Docker
  • 丰富模型: 支持多种开源大模型
  • 完整集成: 与主流 AI 工具无缝对接
  • 本地运行: 保护数据隐私,无需联网

Introduction

Ollama is an open-source platform for running large language models locally. It makes deploying and using models like Kimi-K2.5, GLM-5, MiniMax, DeepSeek, and Gemma incredibly simple. Developers can quickly get started with AI development without complex setup.

Quick Start

Installation

macOS:

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

Linux:

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

Windows:

irm https://ollama.com/install.ps1 | iex

Basic Usage

Launch Ollama:

ollama

Run specific models:

ollama run gemma3

Development Integration

Ollama supports integration with various AI tools:

Claude Code Integration:

ollama launch claude

OpenClaw Assistant:

ollama launch openclaw

Supported integrations include Codex, Copilot CLI, Droid, OpenCode, and more.

REST API

Ollama provides a complete REST API:

curl http://localhost:11434/api/chat -d '{
  "model": "gemma3",
  "messages": [{
    "role": "user",
    "content": "Why is the sky blue?"
  }],
  "stream": false
}'

Advantages

  • Easy to use: Run models with a single command
  • Multi-platform: macOS, Windows, Linux, Docker support
  • Rich models: Supports multiple open-source LLMs
  • Complete integration: Seamless对接 with mainstream AI tools
  • Local execution: Data privacy, no internet required

评论