🦙

Ollama

Ollama is an open-source tool from Ollama Inc. that lets you download and run open-weight large language models — Llama, Mistral, DeepSeek, Gemma, and more — directly on your own machine. It ranks #5 among trending AI tags at roughly 15.3% usage share, driven by developers who want to prototype and run LLMs without paying per-token API fees or sending data to a cloud provider. It bundles a model downloader, a local inference server, and a simple CLI into one command-line install.

Quick facts
Type: Local LLM runtime / model manager
Made by: Ollama Inc.
License/Access: Free and open-source
Language/Platform: Cross-platform — macOS, Linux, Windows; CLI + local REST API
Primary use case: Running open-weight LLMs locally for private, offline, or cost-free inference and development

Jump to: Example Getting started Best for

Example

Once Ollama is installed, running a model is a single command. It also exposes a local REST API on port 11434 that any app can call.

# run an interactive chat with Llama 3 in your terminal
ollama run llama3

# call the same model from code via the local REST API
curl http://localhost:11434/api/generate -d '{
  "model": "llama3",
  "prompt": "Why is the sky blue?",
  "stream": false
}'

Getting started

Install Ollama, pull a model, and start chatting — no account or API key required.

# macOS / Linux install script
curl -fsSL https://ollama.com/install.sh | sh

# or download the Windows/Mac installer from ollama.com/download

# pull and run your first model
ollama run llama3

Best for: Developers who want to experiment with LLMs for free, build offline or privacy-sensitive apps, or run models on their own GPU without recurring API costs — ideal for local prototyping before (or instead of) moving to a hosted API.