> ## Documentation Index
> Fetch the complete documentation index at: https://docs.snapgen.org/llms.txt
> Use this file to discover all available pages before exploring further.

# Account & Billing

> Understand SnapGen prepaid balance, credits, prices, and refunds.

<AccordionGroup>
  <Accordion title="Does SnapGen require a subscription?">
    No. SnapGen uses prepaid USD balance. Add funds in **Console > Billing**
    and pay only for metered requests.
  </Accordion>

  <Accordion title="How do credits map to USD?">
    100 credits equal \$1.00. For example, Seedance 2 costs 100, 200, or 300
    credits for 5, 10, or 15 seconds.
  </Accordion>

  <Accordion title="How are text models billed?">
    Text routes have separate input and output prices per one million tokens.
    The individual model page lists both rates.
  </Accordion>

  <Accordion title="How are image models billed?">
    GPT Image 2 costs $0.01 per successful image, and GPT Image 2 All costs
            $0.02.
  </Accordion>

  <Accordion title="How are video models billed?">
    Gemini Omni and Veo 3.1 have fixed per-request prices. Seedance 2 uses the
    selected 5, 10, or 15-second tier. The Portal estimates the charge before
    submission.
  </Accordion>

  <Accordion title="Are failed video tasks charged?">
    A task that reaches the normalized `failed` state is refunded and reports a
    zero final cost. Check the task and Console logs before resubmitting.
  </Accordion>

  <Accordion title="Where do I see current prices?">
    Use the [live models page](https://snapgen.org/models). Documentation
    pages record current rates, but the live catalog remains the source of truth.
  </Accordion>
</AccordionGroup>