Onboarding

Try orbit for yourself.

Let Orbit summarize articles, emails, videos and more.

Try Orbit on a Video Try Orbit on an Article

How to use

Summarize

Click Summarize to get detailed summaries of emails, articles, documents, and videos automatically.

Ask Orbit

Use Ask Orbit to find specific information or get answers to your questions quickly.

Settings

In Settings, customize your response formats from bullet points, paragraphs, or short sentences.

Orbit supports all major websites

Frequently asked questions

How does Orbit work?

Orbit is a Firefox add on that uses AI to summarize and answer queries about web content such as articles and videos.

When a user asks Orbit to summarize or query content, Orbit gathers the context (eg. text, images, videos, etc.) of the page the user is viewing and provides a summary or answer. Orbit works on websites including Gmail, Wikipedia, NY Times, YouTube, and more.

For the current version, we are using a Mistral LLM (Mistral 7B) hosted within Mozilla’s GCP instance.

Do I need to create an account to use Orbit?

No, you can use Orbit without creating an account.

Does Orbit save the content of the pages I visit or summaries generated?

No, it does not.

When you use Orbit, we receive a payload back that contains the contents of your query; information about the model queried (such as the name and version number); information about technical problems with processing the query, if any; the number of tokens required to process the query; and the model outputs in response to the query. We do not store this data beyond temporarily caching it to process your query and return the outputs to you.

Orbit summaries are only available on the page that you are actually on. As soon as you navigate away from that page, Orbit erases the session.

Is my personal information used to train the AI models?

No, it is not.

The Mistral 7B LLM used to support Orbit is hosted by Mozilla and the queries are never shared with Mistral or any other services.

If you don’t train the models, how will the service improve?

We have built Orbit to remain agnostic about which model we use and we are constantly benchmarking the most recent open-source models available. This allows us to easily swap out models as technology improves.

The tradeoff of not sharing user data with the model is that we cannot influence the model or train it.