Core Concepts

Tokens & Credits

Understanding how Jeanette's usage is measured and billed.

Tokens

Tokens are the unit of AI computation. Every message you send and every response Jeanette generates consumes tokens.

  • Input tokens — the text you send to Jeanette, plus her conversation history and any context she loads (memories, email content, etc.)
  • Output tokens — the text Jeanette writes back to you

A rough guide: 1,000 tokens ≈ 750 words. A typical back-and-forth exchange uses 1,000–3,000 tokens. Long tasks with many tool calls (like a deep research job) can use 50,000 tokens or more.

Your plan includes a monthly token allowance split between input and output. When you reach the limit, Jeanette will let you know and pause until the next billing period.

Credits

Credits are a separate currency used for real-world actions that have direct costs beyond AI computation.

What uses credits

Action Approximate cost
Web search (per query) 1–2 credits
News search (per query) 1–2 credits
Image generation 10–20 credits
Image editing 5–10 credits
Deep research (per section) 5 credits
External API connector calls Varies

What does not use credits

  • Regular conversation
  • Weather queries
  • Email and calendar operations
  • Scheduled task execution
  • Memory operations
  • Background task management

Checking your usage

You can view your current token and credit usage at any time:

You: How many tokens have I used this month?

Jeanette: You've used 1.2M input tokens and 320K output tokens out of your 5M / 2M monthly allowance. You have 380 credits remaining.

Or check the Billing page in the web dashboard for a detailed breakdown.

Topping up credits

On paid plans, you can top up credits at any time from the dashboard without waiting for your next billing period.

When limits are reached

When you reach your token or credit limit, Jeanette will send you a message explaining what happened and how to continue. She won't silently fail — you'll always know why she stopped.

Plans

See the Billing & Plans page for current plan details and token/credit allowances.