Jurat uses simple, transparent per-token pricing. You only pay for what you use — no subscriptions, no lock-in contracts, no surprises.
Tokens are the units that AI models use to process text. You can think of a token as roughly a piece of a word — on average, one token is about three-quarters of a word. A short sentence might be 10–15 tokens, while a full page of text is typically around 400–500 tokens.
When you submit a research query to Jurat, tokens are used to process your question, conduct research, read and analyse material, and generate the response. The total cost of a query depends on how much text is involved — shorter, more focused queries cost less, while longer, more detailed analyses use more tokens.
Jurat offers two different AI models to conduct research queries. A research task conducted using the Normal model costs $6 per million tokens (MTok), and a task using the Advanced model costs $10/MTok.
Jurat uses multiple strategies to save you money during research without reducing the quality of its work. This means that, while Jurat may consider over 30 cases through the course of a single research task, only a portion of those cases will be analysed in depth, reducing the total number of tokens used.
For examples of research outputs along with their pricing, click here.
Traditional legal research tools charge flat monthly fees regardless of how much you use them. That model works for large firms running hundreds of searches a month, but it's prohibitively expensive for solo practitioners, small firms, and legal clinics that only need occasional access.
Per-token pricing means you pay proportionally to the work Jurat does for you. Use it once a week or fifty times a day — your cost always reflects your actual usage.
Jurat Pro Bono is a program that provides our tools at-cost to qualifying non-profit and public interest organizations. This initiative supports the mission of JuratAI to make high-quality AI legal tools available to all practitioners in Canada. For more information, click here.