ollama/ollama

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

What is this?

Ollama is a piece of software that lets you download, run, and manage open-source AI models (like Gemma 4) directly on your own computer or server. Instead of sending your data to a remote cloud service like OpenAI, models run locally on your hardware, giving you complete privacy.

Think of it as a local "server" for AI models. You install Ollama, tell it which model you want, and it handles all the technical details of downloading and running that model—no special AI expertise or expensive cloud accounts needed.

Why is this useful?

  • Privacy: Your conversations and data never leave your machine.
  • Developer-friendly: It exposes a simple REST API (localhost:11434) and official Python and JavaScript libraries, making it easy to add AI to your own apps.
  • Works with your tools: You can plug it into coding assistants (like Claude Code or Codex) to run AI locally, or use it to power personal assistants on messaging apps (via OpenClaw).
  • Huge ecosystem: Because it's so easy to use, hundreds of third-party tools (chat interfaces like Open WebUI, agent frameworks like LangChain and AutoGPT, and RAG systems like RAGFlow) support it out of the box.

How it works

It runs on your machine with three main interfaces:

  1. Command line: Type ollama run gemma4 to chat directly in your terminal.
  2. REST API: Send a curl request to http://localhost:11434/api/chat with your prompt, and get a response back. This is how other software talks to it.
  3. Official SDKs: Install pip install ollama (Python) or npm i ollama (JavaScript) to call models from your own code.

Under the hood, Ollama uses llama.cpp, an efficient open-source project that makes it possible to run large AI models on standard consumer hardware (like your laptop's CPU or GPU) rather than requiring a server full of specialized AI chips.

In short, it's the easiest way to get started building or experimenting with open-source AI models on your own terms.

Related

  • Project
    HelgeSverre/ollama-gui

    A browser‑based single‑page app that connects directly to a local Ollama server, providing chat, model management, history, attachments, markdown rendering, and full offline storage—all without a separate backend.

  • Dispatch
    Ollama Cloud Models Preview

    Ollama has introduced Cloud models in preview, enabling users to run large-scale models on datacenter-grade hardware while maintaining the same local toolset and API integration.

  • Project
    sammcj/gollama

    Gollama is a Go‑based terminal UI for listing, inspecting, editing, running, unloading, copying, deleting and pushing Ollama models, with command‑line shortcuts and a vRAM estimator to help decide if a model fits on your hardware.

  • Dispatch
    llama.cpp and llama.app: Local LLM Inference and Ecosystem

    llama.cpp and its official home llama.app provide high-performance, hardware-agnostic local LLM inference, now featuring a simplified 'llama serve' command and integration with local coding agents like Pi.

  • Project
    ollama/ollama-js

    Official JavaScript client for Ollama, letting Node or browser apps talk to local or cloud Ollama LLM servers. Provides chat, generate, model‑management, embedding, web‑search, and decision‑making APIs with optional streaming and custom host/headers.