What are credits?

MOSTLY AI’s credit system is designed to give you transparency, flexibility, and control so you can create and explore synthetic data with confidence

In short, credits are the currency used by the MOSTLY AI Platform and the MOSTLY AI Asisstant. Anytime you create synthetic data with a generator or ask the Assistant to do something like provide information or generate a visualization, you consume a small number of credits depending on the complexity and size of the request or task.

Users on the Free Plan receive 5 new credits every day, which is enough to power meaningful experimentation and exploration. Heavier usage or more complex tasks may require purchasing additional credits or upgrading to a paid plan.

What are tokens?

Credits are tied to the amount of data processed by the Platform or Assistant, when chatting with the Assistant, this usage is measured in tokens.

1 million tokens = 5.0 credits.

But what exactly is a token? In simple terms, tokens are chunks of data like words, numbers, or values in a table. The Assistant breaks down your input data into tokens so it can analyze and synthesize it effectively. For example, a single row with several fields may be made up of dozens or even hundreds of tokens depending on its contents.

Think of tokens as the raw materials the Assistant uses to generate synthetic versions of your data.

How can I get more credits?

There are a few easy ways to get more credits:

• Daily top-up: Every Free tier user receives 5 new credits each day, automatically.
• Paid plans: Upgrade your workspace to get higher daily limits.

How can I conserve credits?

There are several ways to stretch your usage to ensure that you make the most of your credits:

• Start a new chat for each topic or dataset: This helps the Assistant stay focused and reduces misunderstandings that might lead to costly retries.
• Be precise in your requests: Clear, specific prompts avoid wasted effort and make the Assistant more efficient.
• Use small samples for prototyping: Run test generations on a subset of your dataset before scaling up. You’ll get faster results and save credits.
• Limit unnecessary fields: Clean or trim your input data so the Assistant only processes the most relevant tokens.

By using the Assistant with intent and structure, you’ll get powerful results while keeping your credit usage under control.