> ## Documentation Index
> Fetch the complete documentation index at: https://docs.qbraid.com/llms.txt
> Use this file to discover all available pages before exploring further.
> ## Agent Instructions
> Prefer the qBraid CLI for programmatic platform actions: pip install 'qbraid-cli>=0.12', then run `qbraid configure` once with an API key from https://account.qbraid.com/account/api-keys.
> Always install the latest packages (pip install -U qbraid qbraid-cli); do not pin versions from memory. qbraid-cli below 0.12.0 is incompatible with the current API.
> Device IDs use the QRN format vendor:provider:type:name (e.g. qbraid:qbraid:sim:qir-sv, rigetti:rigetti:qpu:cepheus-1-108q). Legacy underscore IDs are deprecated.
> The REST API base URL is https://api-v2.qbraid.com/api/v1, authenticated with an X-API-Key header.
> Free simulators cost no credits; QPU and GPU jobs consume credits. Surface the estimated cost to the user before submitting a paid job.
> For account signup, API keys, credits, and end-to-end action recipes, see https://qbraid.com/llms.txt.
# Usage Quotas
> Understanding AI usage limits and credit consumption on qBraid
qBraid provides a default monthly quota for AI chats, powered by qBraid credits. This guide explains how your AI quota works and how to track your usage.
## Plan Quotas
Your monthly AI quota depends on your subscription plan:
| Plan | Monthly AI quota | Auto-switch default |
| ------------ | -------------------- | ------------------- |
| **Free** | \$5 (500 credits) | off |
| **Standard** | \$15 (1,500 credits) | on |
| **Pro** | \$50 (5,000 credits) | on |
Auto-switch starts off on the Free plan on purpose: a free user should never
burn credits on AI by surprise. Turn it on yourself if you want messages to
keep working past the quota.
Complex queries with longer responses consume more input / output tokens and
will fill up your quota faster.
## Usage Costs
AI usage is metered by tokens, not by message. Each model has a price per
million input and output tokens, and each message costs the tokens it
actually uses. Three things raise the cost of a message:
* **Longer prompts and longer answers.** More tokens in, more tokens out.
* **Bigger models.** Frontier models price higher per token than fast models.
* **Agent Mode.** Tool calls add context to the conversation, so agent runs
use more tokens than plain chat.
Token costs convert to credits at 100 credits = \$1.
## When the quota runs out
Every message draws from your monthly quota first. What happens next depends
on one toggle, **Auto-switch to Credits**, and your credit balance:
1. **Quota remaining.** The message is covered by your plan. Credits are
untouched.
2. **Quota exhausted, auto-switch on, credits available.** The message bills
your credit balance instead. A message that lands exactly on the boundary
splits: the remainder of the quota covers part, credits cover the rest.
3. **Quota exhausted, auto-switch off.** The request fails with "AI chat
quota exhausted. Enable credit spending to continue." Turn the toggle on
or wait for the monthly reset.
4. **Quota exhausted, auto-switch on, no credits.** The request fails with
"AI chat quota exhausted and no credits available." Buy credits or wait
for the reset.
Two details worth knowing:
* A leftover below one cent counts as exhausted. The panel then shows 100%
used, and the next message already follows the rules above.
* In a team organization, the admin's auto-switch setting wins. A member
cannot spend org credits that an admin has locked, and the usage panel
shows the effective policy, not your own toggle.
### A worked example
A Free-plan user has used 80% of the monthly quota. The usage panel shows
80% used and "can send". Both are true. 20% remains, so the next message is
allowed.
Now the user pastes a long paper and asks `gpt-5.5` for a detailed summary.
At this model's token rates, the message works out to:
* Input: 200,000 tokens = 20% of the monthly quota
* Output: 15,000 tokens = 9%
* Message cost: 29% of the quota, against the 20% that remains
Here is the part that surprises people. The quota check runs before the
message. The cost is only known after the model answers. The remaining 20%
is enough to admit the message, so it runs, and the bill comes to 29%. The
extra 9% has to land somewhere.
With auto-switch on and credits in the wallet, the quota covers its last
20% and the wallet pays the 9% overflow as credits. The message succeeds.
The panel moves to 100% used, and later messages run on credits alone.
With auto-switch off, or with an empty wallet, the request fails. Free
accounts start with auto-switch off, so this is the default experience.
The fix is the same either way: turn on **Auto-switch to Credits** and keep
some credits, or wait for the monthly reset.
The lesson: one long message can cost more than days of short ones. Cost
follows tokens, so a single long paper on a frontier model can outweigh a
week of ordinary chat.
## Tracking Your Usage
Monitor your AI usage in real-time from your account dashboard:
1. Navigate to your [account dashboard](https://account.qbraid.com/)
2. Go to **Account > Usage > AI Chat Usage**
3. View your current consumption and remaining quota
The usage panel displays:
* **Monthly Quota** — Progress bar showing percentage of your monthly AI quota used
* **Status** — Whether you're currently using your quota or credits
* **Renews** — The date when your monthly quota resets
* **Can Send** — Whether you can currently send AI agent messages
* **Auto-switch to Credits** — Toggle to continue chatting with credits when your quota runs out
## Tips for Efficient Usage
Ask multiple related questions in a single prompt rather than separate messages to reduce credit consumption.
{" "}
Save Agent Mode for execution tasks. Use regular AI Chat for exploration and
learning.
The AI remembers your conversation, so you don't need to repeat context in follow-up questions.
## Upgrading Your Plan
Need more AI usage? Visit your [subscription settings](https://account.qbraid.com/account/wallet) to upgrade your plan, or [contact us](mailto:contact@qbraid.com) for enterprise plans with custom AI allocations.
***
For complete details on all qBraid usage limits including compute hours and
disk storage, see [Usage Limits](/v2/lab/user-guide/usage-limits).
