AI usage quotas

AI usage consumes your plan’s allowance before it uses wallet credits. Usage and credit balances belong to the selected workspace.

Plan allowances

These are AI spending allowances, not deposits into your wallet. Check Account > Usage > AI Chat Usage for the actual renewal date and any account-specific allocation. All current tiers can use the model catalog. A more expensive model consumes the same allowance faster.

Token costs

Each model has separate rates for input and output tokens. Models and pricing lists the rate units and explains how to inspect the model picker. Longer prompts, longer answers, and tool results that add conversation context can increase usage.

For an uncached request, estimate the cost in credits as:

(input tokens × input rate + output tokens × output rate) / 1,000,000

Use rates in credits per million tokens. Cached-token billing can change the final cost. qBraid supports fractional credits, so a small request does not automatically round up to a whole credit.

When the allowance runs out

• With allowance remaining, usage draws from the allowance.
• With Auto-switch to Credits enabled, usage beyond the allowance can draw from the wallet. A request that crosses the boundary can use both.
• With auto-switch disabled, exhausted allowance blocks further requests. Enable credit spending if permitted, upgrade, or wait for renewal.
• With auto-switch enabled but no available credits, add credits or wait for renewal.

A remainder of $0.01 or less counts as exhausted when the application decides whether you can send another request. In an organization, the organization’s effective policy can override a member’s personal auto-switch setting.

The final token cost is known after generation. A request can exceed the allowance that remained when it started. The displayed ability to send is not a guarantee that the remaining allowance will cover the entire response. An outstanding AI balance can also block later requests until it can be settled.

Track usage

1. Select your workspace.
2. Open Account > Usage.
3. Inspect AI Chat Usage for consumption, the renewal date, and whether you can send requests.
4. Review Auto-switch to Credits before you continue beyond your allowance.

For individual charges, use Wallet payment history. Organization administrators can inspect organization transactions.

Reduce costs

Select a model with a lower token rate when it meets your needs. Keep prompts and requested output focused. Use agent tools when the task needs them, since tool results add context that the model must process.

To increase your allowance, compare personal subscriptions or team subscriptions. For compute and storage limits, see Usage limits.