Pricing

Transparent credit-based pricing with clear spend caps — start small, upgrade anytime.

Full-power pool for coding agents, IDEs and heavy workloads.

Daily Trial

Daily

For new users who want to try before moving to a weekly/monthly plan

5credits/day

What's included

Daily limit
250 credits
Get started

Weekly Basic

Weekly

For steady, light users

12credits/wk

What's included

Daily limit
250 credits
Weekly limit
1,200 credits
Get started

Weekly Pro

Weekly

For heavy weekly users

24credits/wk

What's included

Daily limit
500 credits
Weekly limit
2,500 credits
Get started
Most popular

Monthly Starter

Monthly

For light daily users

39credits/mo

What's included

Daily limit
150 credits
Weekly limit
800 credits
Get started

Monthly Plus

Monthly

For everyday users

49credits/mo

What's included

Daily limit
300 credits
Weekly limit
1,500 credits
Get started

Monthly Pro

Monthly

For power users with heavy daily use

89credits/mo

What's included

Daily limit
500 credits
Weekly limit
2,500 credits
Get started

Max

Monthly

Highest retail plan — for whales/small teams

199credits/mo

What's included

Daily limit
1,000 credits
Weekly limit
5,000 credits
Get started

Custom

Contact sales

Need a custom plan? Talk to our sales team.

Contact sales

Pricing#

MintRouter bills the way the model providers do: in US dollars per 1 million tokens (USD / 1M tokens). There are no hidden seat fees and no currency conversion to reason about — the price you see is the price you pay for the tokens you send and receive.

How billing works#

Every request you route through MintRouter is metered for two things:

  • Input tokens — the prompt, system message, and any context you send.
  • Output tokens — everything the model generates back to you.

Each model has its own input and output rate, quoted in USD / 1M tokens. Your cost for a request is simply:

(input_tokens  / 1,000,000 × input_rate)
+ (output_tokens / 1,000,000 × output_rate)

Because rates are per-million, a typical short request costs a fraction of a cent. You can estimate any request's cost up front from its token counts.

Plans and quotas#

A plan gives you access to the router plus spending quotas that protect you from runaway bills:

  • Daily cap — the most you can spend in a single day.
  • Weekly cap — a rolling ceiling across the week.

When a cap is reached, further requests are paused until the window resets, so a loop or a misconfigured client can never quietly drain your balance. Higher plans raise the caps and allow more concurrent requests.

What's included#

  • Access to every model MintRouter routes to, through one compatible endpoint.
  • Per-request usage metering with full history in your dashboard.
  • Daily and weekly quota enforcement.
  • OpenAI- and Anthropic-compatible APIs at no extra charge.

FAQ#

Do you charge per seat or per request? No. You pay for tokens. Plans set your quotas and concurrency, not a per-seat fee.

How do I know what a model costs? Each model lists its input and output rate in USD / 1M tokens. Multiply by your token counts to estimate cost before you send.

What happens when I hit my cap? Requests are paused until the daily or weekly window resets. Your data and keys are untouched — only new spending is held.

Can I change plans later? Yes. Plans are designed to scale with you; raise your quotas as your usage grows.

Is unused quota refunded? Quotas are spending ceilings, not prepaid balances — see the Policy page for the full refund summary.