Pricing, Balance, and Quota

NexoRouter uses prepaid balance and usage-based billing. You add balance first, then API requests consume quota according to model usage.

The Billing dashboard screenshot uses sanitized demo data. Your real balance, quota, transactions, and receipts will differ.

Core concepts

Top-up options

The Dashboard offers standard top-up packages:

Custom top-ups are supported from $1 to $10000, rounded to cents.

How requests are charged

Most chat requests use:

• input tokens;
• output tokens.

Each model has its own input and output price. The actual charge is visible in Usage Logs as both USD and quota.

Balance and key budgets

Workspace balance is shared, but key budgets let you limit risk.

Recommended setup:

Insufficient balance

insufficient_quota means the workspace balance or key budget cannot cover the request.

Fix it:

1. Open Billing and check current balance.
2. Open Usage Logs and check recent spend.
3. Add balance or create a replacement key with the right budget.
4. Retry the request.

Cost control

• Create separate keys for dev, staging, and production.
• Use hard key budgets for experiments and external tools.
• Start with lower-cost models during integration testing.
• Set max_tokens during development.
• Avoid automatic infinite retries.
• For slow models, raise the client timeout instead of canceling and retrying quickly.

Payment boundaries

Payment support follows the product paths available in the Dashboard:

• Top-ups are made from Billing.
• Stripe Checkout is the supported online checkout path.
• Balance is credited after the signed payment confirmation is processed.
• Refunds, disputes, and manual credits require human review.