What is an LLM? Large Language Models Explained Simply

Glossary › AI Computing › LLM

What is an LLM? (Large Language Model)

An LLM is the type of AI behind assistants like ChatGPT and Claude — a model trained on vast amounts of text that can write, summarise, code and answer questions.

Running LLMs on your own PC

Open models like Llama, Mistral, Qwen and Gemma can run locally on your own machine using free tools like Ollama or LM Studio — completely private, no subscription, works offline. Model “size” is measured in parameters (7B, 13B, 70B) — bigger is smarter but needs more VRAM.

What hardware do you need?

An 8GB NVIDIA GPU runs 7B–8B models well — the same RTX cards in our gaming builds. 12GB handles 13B models; 16GB runs larger quantised models. NVIDIA cards are strongly preferred because most AI tooling is built on CUDA. See which RTX cards have 12GB or 16GB in our RTX GPU comparison.

Common questions

Can I run an LLM locally without a GPU?

Yes. Small models run on the CPU using normal RAM, with tools such as Ollama or LM Studio. It is much slower than on a graphics card, often a few words per second, so it suits small models and light use.

How much RAM do I need to run an LLM locally?

8GB will run only the smallest models. 16GB is a practical minimum for 7B models, and 32GB gives room for larger ones. With a graphics card, VRAM matters more than RAM: a 7B model needs about 5 to 6GB of VRAM.

Can I run an LLM locally on Windows?

Yes. Ollama and LM Studio both install on an up-to-date Windows 10 or 11 PC like any other program and can use an NVIDIA graphics card automatically. No Linux or coding is needed to get started.

More questions? See all common PC questions.

Related terms

VRAMCUDAGPU

💡 Want your own private ChatGPT at home? Our RTX builds run local AI — WhatsApp us.