> For the complete documentation index, see [llms.txt](https://references.everstrike.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://references.everstrike.io/ai/events.md).

# Events

Inspect AI strategy decisions, event details, and evaluation costs.

Active [strategies](/ai/strategies.md) produce events. Events record each AI evaluation and help you review decisions, refine prompts, and track credit use.

Each active strategy produces one event per evaluation for each selected market. Paused strategies produce no events, place no trades, and consume no AI credits.

{% hint style="info" %}
Use events to validate a new strategy before increasing its trade quantity or evaluation frequency.
{% endhint %}

### Review an event

1. Check **Action**, **Amount**, and **Order Submitted** to confirm the outcome.
2. Compare **Confidence** and **Reasoning** with the strategy's prompt.
3. Review **Credits Used** and token counts when managing costs.

### Event properties

| Property          | Description                                                                         |
| ----------------- | ----------------------------------------------------------------------------------- |
| Creation Time     | Time when the event was first recorded.                                             |
| Pair              | Market evaluated by the strategy.                                                   |
| Action            | AI decision: `BUY`, `SELL`, `HOLD`, or `CLOSE`.                                     |
| Amount            | USD value of orders submitted for the event.                                        |
| Price             | Market mark price at creation time.                                                 |
| Confidence        | AI confidence in the decision, from 0% to 100%.                                     |
| Reasoning         | Reasoning that led to the selected action.                                          |
| Strategy          | Strategy that generated the event.                                                  |
| Order Submitted   | Whether the event submitted an order.                                               |
| Forced Evaluation | Whether the event came from a forced evaluation.                                    |
| Model             | AI model used for the event.                                                        |
| Prompt Tokens     | Tokens consumed by the prompt.                                                      |
| Completion Tokens | Tokens consumed by the model response.                                              |
| Discretion        | Whether the AI used discretion.                                                     |
| Cost Savings Mode | Whether the event used Cost Savings Mode. Reasoning output is minimal in this mode. |
| Thinking Time     | Time the AI took to reach its decision.                                             |
| Credits Used      | AI credits consumed by the event.                                                   |

### Event frequency and credits

Evaluation frequency determines how quickly events accumulate. Every event consumes AI credits, which are automatically deducted from your USD balance.

Pause a strategy to stop event generation, trading, and credit use. You can resume a paused strategy at any time.

#### GPT-4.1-mini estimates

These estimates apply to each selected market.

| Evaluation frequency | Events per market per day | Estimated credits per market per hour |
| -------------------- | ------------------------: | ------------------------------------: |
| 1s                   |                    86,400 |                                 $0.54 |
| 5s                   |                    17,280 |                                $0.108 |
| 15s                  |                     5,760 |                                $0.036 |
| 30s                  |                     2,880 |                                $0.018 |
| 1m                   |                     1,440 |                                $0.009 |
| 3m                   |                       480 |                                $0.003 |
| 5m                   |                       288 |                               $0.0018 |
| 15m                  |                        96 |                               $0.0006 |
| 30m                  |                        48 |                               $0.0003 |
| 1h                   |                        24 |                              $0.00015 |

### Event Costs

Costs vary by the model used for the event.

| Model                        | Cost per event |
| ---------------------------- | -------------: |
| GPT-4o-mini                  |      $0.000125 |
| GPT-4.1                      |        $0.0015 |
| GPT-4.1-mini                 |       $0.00015 |
| GPT-4.1-nano                 |      $0.000075 |
| GPT-5-mini-minimal-reasoning |        $0.0002 |
| GPT-5-nano-minimal-reasoning |       $0.00005 |
| GPT-5-mini-low-reasoning     |        $0.0004 |
| GPT-5-nano-low-reasoning     |        $0.0001 |
| GPT-5-mini-medium-reasoning  |        $0.0008 |
| GPT-5-nano-medium-reasoning  |        $0.0002 |
| GPT-5-mini-high-reasoning    |         $0.002 |
| GPT-5-nano-high-reasoning    |        $0.0004 |


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://references.everstrike.io/ai/events.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.