Ollama

Actively maintained

Run large language models locally

AI
Stars
★ 178k
Version
v0.32.6
Last commit
1 day, 17 hours ago
License
MIT
Language
Go
Setup
Easy
Min RAM
8192 MB

Metrics verified from the source repo on Aug 7, 2026.

What it is

The easiest way to download and run open large language models locally on your own hardware — a private, offline alternative to hosted LLM APIs.

Replaces

OpenAI API

Self-host Ollama instead of paying for the above.

Highlights

  • One-command local model runs
  • OpenAI-compatible API
  • Runs Llama, Mistral, Qwen & more

How to deploy

DockerDocker ComposeBare metal

Recent changes

  • Aug 6 version Version: v0.32.5 → v0.32.6
  • Jul 28 version Version: v0.32.4 → v0.32.5
  • Jul 26 version Version: v0.32.3 → v0.32.4
  • Jul 24 version Version: v0.32.1 → v0.32.3

More ai apps

n8nWorkflow automation for technical teams ★ 200k Open WebUIExtensible, self-hosted AI chat interface ★ 148k ImmichSelf-hosted photo and video backup ★ 110k nanobotAn ultra-lightweight, self-hosted personal AI a… ★ 47k