Back to Directory

Ollama

Run Llama 3, Mistral, Gemma, and other models locally.

WebsiteInstall

Overview

Ollama is a lightweight, extensible framework for building and running language models on your local machine. It provides a simple CLI and a local API that is compatible with many tools.

Capabilities & Support

text
vision
audio
image
video

Popular Models

Model NameTypeSizeBest For
Gemma 4 E4B
textvisionaudio
5.6 GBGoogle's April 2026 sweet spot. Native Audio ASR/Translation + Vision. Perfect for M4 MacBooks.
Qwen 3.6-Plus
textvision
24BAlibaba's April 2026 flagship. State-of-the-art coding and multimodal understanding.
Llama 4 Maverick
text
80BMeta's legendary reasoning model (2025). Highly reliable for complex agentic workflows.
DeepSeek V4-Lite
text
16BFast reasoning model from the DeepSeek family, optimized for 2026 hardware.
Gemma 4 E2B
textvisionaudio
3.6 GBUltra-fast multimodal inference. Runs perfectly on 16GB RAM devices.

Installation Guide

1

Download & Install

Go to ollama.com and download the installer for your OS. Run the installer and finish the setup.

https://ollama.com/download
2

Verify Installation

Open your terminal and run the ollama command to see if it's working.

ollama --version
3

Run your first model

Download and run the Llama 3.2 model directly from the CLI.

ollama run llama3.2