📚 AI Resources

Tools, models, and guides for your AI journey

🛠️ Recommended Tools

🦙

Ollama

Run Llama, Mistral, Qwen, and more locally

Ollama is the easiest way to run large language models locally. It handles model downloading, setup, and provides a simple command-line interface for chatting with models. Works on Mac, Windows, and Linux.

✓ One-line install

✓ 100+ models available

✓ Automatic GPU detection

✓ API compatible with OpenAI

✓ Free & open source

✓ Works offline

Quick Install

Download from ollama.ai (Mac/Windows) or run curl -fsSL https://ollama.ai/install.sh | sh (Linux)

Run ollama pull llama3.2 to download your first model

Run ollama run llama3.2 to start chatting

🎨

LM Studio

Beautiful GUI for running local models

Download Docs

LM Studio provides a graphical interface for discovering, downloading, and running local LLMs. Perfect for users who prefer clicking over command lines. Includes a built-in chat interface and lets you adjust model settings like temperature and context length.

✓ Visual model browser

✓ Built-in chat interface

✓ Model parameter controls

✓ Local server mode

✓ Mac & Windows

✓ Free to use

Quick Install

Download from lmstudio.ai

Open the app and browse models in the sidebar

Click a model to download, then start chatting

✨

OpenClaw

Full AI assistant with tools & automation

Download Setup Guide

OpenClaw is a desktop AI assistant that builds on Ollama to provide a complete AI experience. It includes local and cloud model support, file analysis, web browsing, code execution, and automation features. Designed for productivity and privacy.

✓ Local & cloud models

✓ File & code analysis

✓ Web browsing

✓ Automation & skills

✓ Memory & context

✓ Mac-first

🤗

Hugging Face

Model hub with 500k+ models

Browse Models Docs

Hugging Face is the GitHub of AI models. Browse hundreds of thousands of open-source models for text, image, audio, and more. Most are free to download and use. Great for finding specialized models or exploring what's possible.

✓ 500k+ models

✓ Free to download

✓ Model benchmarks

✓ Community & docs

📦 Recommended Models

Run these locally with Ollama or LM Studio. All are free and open-source.

Model	Size	Best For	Install Command
Llama 3.2 3B	~2GB	General use, balanced performance	`ollama pull llama3.2`
Llama 3.2 1B	~1.3GB	Fast responses, low resources	`ollama pull llama3.2:1b`
Phi-4 Mini	~2.5GB	Reasoning, coding, math	`ollama pull phi4-mini`
Mistral 7B	~4GB	Efficient, fast, great all-around	`ollama pull mistral`
Qwen 2.5 7B	~4.7GB	Coding, multilingual, long context	`ollama pull qwen2.5`
Gemma 3 4B	~3GB	Google's efficient open model	`ollama pull gemma3:4b`
DeepSeek R1	~4.7GB	Advanced reasoning, math, code	`ollama pull deepseek-r1`
Codellama	~4GB	Code generation & completion	`ollama pull codellama`

Tip: Start with llama3.2:1b if you're unsure or have limited RAM. It's small and fast. Move up to larger models as needed.

💻 Hardware Requirements

Minimum

4GB RAM
2 CPU cores
5GB disk space
1-2GB models only
May be slow

Ideal

16GB+ RAM
Apple Silicon or NVIDIA GPU
20GB+ disk space
All models available
Fast responses

🍎 Apple Silicon (M1/M2/M3/M4): Macs with Apple Silicon run local models very efficiently thanks to unified memory. A base M1 Mac mini with 8GB RAM can run most 3-4GB models smoothly.

🖥️ Windows/Linux with NVIDIA GPU: NVIDIA GPUs with 6GB+ VRAM provide excellent performance. RTX 3060 or better recommended. CPU-only works but is slower.

☁️ Cloud Services

No install required, but data goes to servers. Good for complex tasks or when you don't have hardware for local models.

ChatGPT

🚀 Quick Start Guide

Fastest way to get started:

Download Ollama from ollama.ai

Mac: drag to Applications. Windows: run installer. Linux: curl command.

Open Terminal and run:

ollama pull llama3.2

Start chatting:

ollama run llama3.2

That's it! You now have a fully local AI assistant running on your computer. No internet required, completely private.