> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs-staging.joytokens.ai/llms.txt.

# Usage

> Workspace usage and performance analytics

Workspace usage and performance analytics refresh every 5 minutes. Filter the **Overview** and **Usage Details** tabs by Time Range and API Key.

## Overview

Metrics include Credits Used, Total Tokens, Latency, Cache Hit Rate, and Total Requests. Charts cover token usage, p95 latency, cache savings, Top API Keys, Top Members, and Top Models.

![Enterprise developer usage analytics](/_fern-files/_img/d3a5c73bd92ba39dfe7e9d02e64f8a73607b4a880efddddbeb8697d65baa72c1.webp)![](/_fern-files/_img/2507d25be2bd21bf6fe2221b6d0d3590d30411d80915def4f1377a80b66f55fc.webp)

## Usage Details

Switch to the **Usage Details** tab to inspect each individual API call.

**Top filters**

* **Time Range**: defaults to Last 30 days; switchable to Last 7 days or Last 24 hours.
* **API Key**: defaults to All API Keys; filter by a specific key.

**Usage record table columns**

* **Time**: when the call was made (MM-DD HH:mm).
* **User**: the caller's username.
* **API Key**: the key name/alias used.
* **Model**: the model called (partially masked, e.g. `Te***-Embedding-3-Large`).
* **Input Tokens**: input tokens consumed by the request.
* **Output Tokens**: output tokens returned by the model.
* **Cache Read**: tokens read from cache hits.
* **Credits**: the cost/credits deducted for this call.
* **Latency**: request duration in seconds.
* **Status**: call result status (e.g. `POSTED` for recorded).

![Usage details](/_fern-files/_img/ed8b8ec693146069a360bd9bc53e159520ed8b5c4b469effef7532212a4c01a9.webp)![](/_fern-files/_img/c204e11cdea348a8d33685703ac9c5907d2f0e2eb080f52cebc6fc956a39aaab.webp)