Skip to content

Glossary

What is an open-weights model?

An open-weights model is an AI model whose trained parameters are published, so anyone can download and run it on their own hardware under its license.

What it means

Hosted models such as Claude or GPT run only on their provider's servers. Open-weights models, such as Gemma, Llama, Mistral, Qwen and DeepSeek models, publish their weights, so you can run them on your own servers or a cloud of your choice.

Open weights is not always the same as open source. Licenses vary, and some add conditions for large companies or particular uses.

Why it matters for a business

Open weights give full control over data and versions, and can lower cost at very high volume. They also mean you run and maintain the infrastructure yourself, which is a real cost for small teams.

A business example

Things to watch

  • Read the license before building on a model.

  • Include GPU, engineering and monitoring in cost estimates.

  • Compare quality with a hosted model on your tasks.

  • Plan how you will update to newer versions.

Keep exploring

FAQ

Questions about open-weights model

Have a question that is not here? Ask us directly.

Start a project

Tell us what you want to build. We will show you a faster path.

Send a short brief. We reply with questions, a suggested plan and an estimate you can compare with other offers.

Your privacy choices

We use necessary storage to run this site. With your permission we also use Google Analytics to see which pages help people, and load maps from Google. You can change this at any time. Read the cookie policy.