> For the complete documentation index, see [llms.txt](https://docs.revenium.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.revenium.io/track-and-control-costs/analyze-ai-tooling-spend/setup-chatgpt-enterprise.md).

# Setup ChatGPT Enterprise

Connect your ChatGPT Enterprise workspace to see what your team's ChatGPT usage costs you, Codex included, when you pay for it through a ChatGPT Enterprise plan. Revenium reads the workspace's usage per person, per day and per product (ChatGPT on the web, desktop and mobile, agents, Codex and any other client OpenAI reports), in OpenAI credits and in US dollars, and shows it on **Providers** beside the rest of your AI spend.

This is a billing connection, set up once for the whole workspace. Nothing is installed on developers' machines. It is separate from [Setup Codex CLI](https://docs.revenium.io/track-and-control-costs/analyze-ai-tooling-spend/setup-codex.md), which sends per-request telemetry from each developer's Codex CLI. You can use either one, or both: the connection gives the figures OpenAI reports for the workspace, and the CLI setup adds per-request detail.

{% hint style="warning" %}
**An OpenAI API Platform admin key will not work here.** A key from platform.openai.com reads your API organization, not your ChatGPT workspace, and it does not include ChatGPT Enterprise or Codex plan usage. Connect API spend with the **OpenAI** provider instead. See [Provider Integrations](https://docs.revenium.io/integrations/provider-integrations.md#openai).
{% endhint %}

## Requirements

* A ChatGPT Enterprise workspace
* A workspace owner or admin who can create admin keys in the OpenAI Admin Console
* Revenium account with access to **Integrations > Providers**

## 1. Create a workspace admin key

1. A workspace owner or admin opens the OpenAI Admin Console at <https://admin.openai.com/credentials?tab=admin-keys> (**Credentials > Admin keys**).
2. Create a new admin key, choose your ChatGPT Enterprise workspace, and pick **Custom permissions**.
3. Set **Enterprise analytics usage** to **Read**. Also set **Codex analytics** to **Read** if you want Revenium to import up to 120 days of Codex history when you connect.
4. Copy the key. OpenAI shows it only once.
5. Copy the workspace ID. It is shown in the OpenAI Admin Console and looks like `xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx`.

## 2. Add ChatGPT Enterprise as a provider

1. In the Revenium dashboard, go to **Integrations > Providers** and select **+ Link Provider**.
2. Select **ChatGPT Enterprise**.
3. Enter the **Workspace ID**.
4. Paste the key into **Workspace Admin Key**.
5. Select **Create**.

Revenium checks the key against your workspace when you save. If OpenAI refuses it, Revenium shows why, for example when the key is an API Platform admin key or is not authorized for that workspace.

Connect each ChatGPT Enterprise workspace once. To replace a key later, edit the connection and enter the new key together with the workspace ID.

## 3. Verify data is flowing

Saving the connection starts an import. It reads OpenAI's daily usage report as far back as OpenAI serves it. With **Codex analytics** set to **Read**, it fills older days, up to 120 days back, from Codex history, which covers Codex only.

Open **Intelligence > Providers** and select ChatGPT Enterprise, or keep all providers selected. The **ChatGPT Enterprise: OpenAI credits and USD per day** table shows each day's figures in three columns:

* **OpenAI credits**: the OpenAI credits the workspace used that day, across every ChatGPT client.
* **USD, credit-priced usage**: OpenAI's estimate in US dollars of those credits, at the workspace's overage rate. A day with credits but no estimate from OpenAI says **No USD estimate reported**.
* **USD, USD-priced usage**: usage that OpenAI priced directly in US dollars.

Days filled from Codex history show Codex only. The table shows only when the usage type filter is set to all usage types.

## What the connection covers

* **Every ChatGPT client, per product.** Chat on the web, desktop and mobile, agents, Codex and any client OpenAI adds later are each stored as their own product. Days filled from Codex history, older than what OpenAI's daily usage report serves, cover Codex only.
* **Seat fees are not included.** The US dollar figures cover usage, not the fee you pay per seat.
* **Per person.** Usage is stored for each person OpenAI names. Usage the workspace reports without naming a person is kept as workspace spend not attributed to anyone, so each day still adds up to the workspace's own total.
* **Seat coverage.** **Providers > Spend by provider** shows a ChatGPT Enterprise row in **Invoiced vs Metered** that compares your Codex telemetry with the usage on this workspace's invoice, and counts the people billed through it, including people who use ChatGPT without Codex.
* **Recent days can change.** OpenAI can revise recent figures, so Revenium reads the last 30 days again each day. A day's figure is final once it is more than 30 days old. See [Data freshness](https://docs.revenium.io/track-and-control-costs/data-freshness-and-provisional-figures.md).

## Troubleshooting

### The key is refused when you save

* Check that the key was created in the OpenAI Admin Console at admin.openai.com, not at platform.openai.com.
* Check that the key was created for this workspace and that **Enterprise analytics usage** is set to **Read**.
* Check that the workspace ID is the one for the same workspace as the key.

### Less history than expected after connecting

The full 120-day Codex history needs **Codex analytics** set to **Read** on the key. Create a new key with that permission and replace the saved one.


---

# 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://docs.revenium.io/track-and-control-costs/analyze-ai-tooling-spend/setup-chatgpt-enterprise.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.
