> For the complete documentation index, see [llms.txt](https://notara.gitbook.io/notara-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://notara.gitbook.io/notara-docs/billing/credits.md).

# Credits

Credits are how Notara measures and limits LLM usage. One credit equals 10,000 tokens. The Pro plan includes 2,500 credits (25 million tokens) per billing period.

## How Credits Are Counted

Every time the Notara agent makes a call to your BYOK LLM, the input and output tokens are summed and converted to credits:

```
credits_used = (input_tokens + output_tokens) / 10,000
```

For example:

* A brief question and answer: \~3,000 tokens = 0.3 credits
* A medium-length research task with 3 tool calls: \~20,000 tokens = 2 credits
* A long automation run (weekly digest with many tools): \~40,000 tokens = 4 credits
* An unusually long session with deep context: \~100,000 tokens = 10 credits

Credits are tracked at the sub-credit level (e.g., 0.3 credits), so you don't lose partial credits.

## What Counts Toward Credits

| Counts                                           | Does Not Count                                           |
| ------------------------------------------------ | -------------------------------------------------------- |
| Primary agent model calls (Claude, GPT-4o, etc.) | Triage/classification calls (lightweight routing models) |
| Long context calls with many tool results        | Cached context (prompt caching offsets counted net)      |
| Skill prompt injections as part of agent turns   | Dashboard API calls                                      |
| Automation runs                                  | Webhook payload processing overhead                      |

Triage calls use small, cheap models (e.g., `gpt-4o-mini`) to route and classify messages. These are excluded from credit counting because they're Notara's infrastructure cost, not your team's LLM usage.

## Viewing Credit Usage

The dashboard **Overview** page shows a credit gauge with your current usage vs your monthly allowance. It updates with a short delay after each agent turn.

For detailed usage:

1. Go to **Settings → Billing → Usage**.
2. See a day-by-day breakdown of credit consumption.
3. Filter by channel or automation to identify heavy usage.
4. Export the usage CSV if you need to analyze it further.

## Credit Efficiency Tips

If you're using credits faster than expected, here are the highest-impact optimizations:

### 1. Trim Spine Documents

Spine documents are injected on every agent turn. A 10,000-token spine document adds 10,000 tokens to every request. If a document is only relevant occasionally, change it to `on_demand` — it'll only be loaded when semantically relevant.

### 2. Reduce Conversation History

Long Slack threads mean long conversation history. If a thread is growing very long and the agent is referencing old messages, start a new thread to reset the context window.

### 3. Scope Channels

If a channel isn't actively using the agent, remove the bot from it. Channels with the bot active but rarely used still consume minimal credits, but it adds up.

### 4. Efficient Skill Prompts

Verbose skill prompts consume tokens on every invocation. Keep prompts focused — 200-500 words is usually sufficient. Avoid repeating information that's already in your spine documents.

### 5. Check Automations

Scheduled automations run without human prompting and can consume significant credits if they're running frequently or pulling large contexts. Review automation run history to spot inefficient ones.

## Credit Alerts

Notara sends Slack DMs to workspace admins at two thresholds:

* **80% used** — a heads-up that you're approaching the limit
* **100% used** — the limit has been reached; the agent is paused

These alerts go to all admins in the workspace via the Notara bot's DM.

## The Pro Plan in Practice

With 2,500 credits (25 million tokens) per month, here's what a typical active team might consume:

| Activity                                     | Credits/Week | Notes                   |
| -------------------------------------------- | ------------ | ----------------------- |
| 50 Slack conversations/day at 3 credits each | 1,050        | Active team usage       |
| 5 weekly automations at 4 credits each       | 20           | Weekly digests, reports |
| 2 research requests/day at 2 credits each    | 28           | Perplexity searches     |
| **Weekly total**                             | **\~1,098**  |                         |
| **Monthly total**                            | **\~4,390**  |                         |

A highly active team with this usage pattern would exceed the Pro plan's 2,500 credits. More typical teams with lighter Slack usage (10-20 conversations/day) would stay well within the limit. The usage breakdown in **Settings → Billing** will show exactly where your credits are going.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://notara.gitbook.io/notara-docs/billing/credits.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.