Flagship tool: memory calculator
Estimate VRAM and RAM for supported local models using model size, quantisation and context length.
Open the LLM memory calculator →RUN AI ON YOUR OWN COMPUTER
Local AI, without the guesswork.
Choose a runtime, pick a model your hardware can actually hold, and understand which parts of the workflow stay local.
What is Ollama?
Understand the runtime, model library, local API and where it fits.
ExploreCan my PC run local AI?
Check VRAM, RAM and realistic starting points before downloading a model.
ExploreHow much VRAM do you need?
Match model size, quantisation and context length to available GPU memory.
ExploreCan local AI really be private?
Separate local model inference from web search, cloud APIs and remote tools.
ExploreChoose how you want to run models
OLLAMA
Simple runtime and API
Pull models by name, run them locally and expose an API for other tools.
Understand Ollama →LM STUDIODesktop model manager
Search, download, load and chat with local models through a graphical desktop app.
Compare the two →OPEN WEBUIBrowser interface for local or hosted models
Put a polished web interface in front of Ollama or other compatible providers.
Explore Open WebUI →What do you actually want to do with it?
USE CASES
What is local AI actually good for?
Writing, coding, private documents and repeatable local workflows, with limits made clear.
Explore practical uses →DOCUMENTSWork with private documents locally
Understand full-context chats, retrieval and where documents stay on your machine.
Explore local document AI →CODINGUse local models for code
Pick realistic coding tasks and keep testing in the loop.
Explore local coding →When something feels wrong
Local or hosted?
If you are still deciding whether local AI is worth the setup, compare it with hosted assistants and rented GPUs.
Compare local and hosted AI →