Local AI Models

When you use an AI like Lumen or ChatGPT, your question travels to a massive data centre on the other side of the world.

There is another way to run LLMs right on your own device. These are called local models.

To run a model locally you need to be able to download it. This is why all local models are also open weight.

Not all open source models are local; you can run an open weight model on a cloud server.

Llama is the most famous local LLM but there is also Gemma by Google, Phi by Microsoft and Mistral and DeepSeek.