Today, Vantage is launching support for OpenAI Codex costs, expanding Vantage's existing OpenAI integration beyond OpenAI Platform (developer API) costs to cover Codex spend and usage across clients like Codex CLI, Codex Web, Codex VS Code, and Codex GitHub. Vantage now ingests daily estimated cost and token usage from OpenAI's Codex Analytics API, surfacing it in Cost Reports alongside every other provider and broken out by Codex client, model, user, token type, and speed.

Previously, Vantage's OpenAI integration connected only to the OpenAI Platform Costs API, which reports programmatic (developer) API spend. Codex usage, across the CLI, web, IDE extensions, and other clients, is not exposed by that API and therefore could not be tracked in Vantage. To understand Codex consumption, finance and engineering teams relied on the Codex analytics dashboard in the ChatGPT admin console, which sat outside the cost tooling that already covers AWS, GCP, Azure, and the rest of their stack. This made it difficult to build a complete view of AI spend, allocate Codex costs to teams, or apply consistent budgets and alerts.

Now, OpenAI Enterprise customers can connect a ChatGPT workspace to Vantage with a dedicated Codex Analytics API key and workspace ID in a new OpenAI Codex section of the OpenAI integration page. Vantage imports daily estimated cost and token usage into Cost Reports, where customers can analyze spend by Codex client, model, token type, speed, or user; allocate costs with Virtual Tags; and monitor spend with Budgets and Anomaly Detection. Client- and token-type dollar amounts are allocated estimates that reconcile to OpenAI's imported model-level estimated costs.

Support for OpenAI Codex costs is available today for Vantage customers with an OpenAI Enterprise account. To get started, add a Codex Analytics API key in the OpenAI Codex section of the OpenAI Integrations page. See the Vantage documentation for setup details.

Frequently Asked Questions

1. What is being launched today?

Vantage is launching OpenAI Codex cost and token usage support. OpenAI Enterprise customers can connect a ChatGPT workspace using a dedicated Codex Analytics API key and workspace ID, enabling daily ingestion of estimated cost and token usage data into Vantage.

2. Who has access to this integration?

The OpenAI Codex integration is available to Vantage customers with an OpenAI Enterprise account and a ChatGPT Enterprise workspace that has Codex Analytics API access. The customer must be able to provision a dedicated API key for Codex Analytics whose Platform organization matches the organization associated with the ChatGPT workspace.

3. How much does this integration cost?

There is no additional fee for using the integration. However, Codex costs will be included in quota tier enforcement. If your Codex costs push you over your current tier limit, you may be prompted to upgrade. See the Pricing page for details.

4. How does the integration work?

Vantage uses a dedicated Codex Analytics API key and ChatGPT workspace ID to ingest daily, aggregated estimated cost and token usage data from OpenAI's Codex Analytics API. Per OpenAI's documentation, Analytics API results are scoped to a ChatGPT workspace, while requests authenticate with a Platform organization API key whose organization must match the workspace's associated organization. Vantage imports both per-user daily results and workspace daily totals, uses the workspace totals for reconciliation, and presents the transformed data in Cost Reports.

5. What permissions are required on the OpenAI side?

You need a dedicated API key provisioned for Codex Analytics, scoped to codex.enterprise.analytics.read, and the workspace ID from the ChatGPT Admin console. The key's Platform organization must match the organization associated with the ChatGPT workspace. It is best practice to use a dedicated key for Vantage, store it in your organization's secret-management system, and rotate it according to your security policy.

6. What permissions are needed within Vantage?

You must have a Vantage Owner or Integration Owner role to add or remove the OpenAI Codex integration.

7. What permissions does Vantage have in my OpenAI account?

Vantage has read-only access to aggregated Codex analytics data for the connected workspace. The imported daily aggregates can include user email and user ID, which Vantage uses for Resource and tag-based reporting. The API does not provide Vantage with prompts, generated code, or raw audit-event content. Vantage cannot run Codex tasks, change workspace settings, or perform actions that incur costs.

8. What dimensions can Codex costs be filtered and grouped by?

You can filter and group by:

• Billing Account (the ChatGPT workspace ID)
• Account (the account name you set when connecting)
• Service (the Codex client the spend came from, such as Codex CLI, Codex Web, or Codex VS Code)
• Category (the model, such as gpt-5.1-codex)
• Subcategory (uncached input, cached input, and output tokens)
• Resource (the user the spend is attributed to, by email)
• Charge Type (usage)

The following provider tags are also available to filter, group, and build Virtual Tags on:

• openai:billing_source — the data source the row came from (admin for OpenAI Platform, codex_analytics for Codex)
• openai:client_id — the Codex client
• openai:model — the model
• openai:speed — the service speed tier
• openai:user_email — the user's email
• openai:user_id — the user's ID
• openai:cost_attribution — set to unattributed on reconciliation rows for spend not attributed to a specific user

Client- and token-type dollar amounts are derived allocations. Model-level estimated cost and speed originate from OpenAI's model breakdown.

9. Can I view usage data for OpenAI Codex?

Yes. Vantage reports uncached input, cached input, and output token usage. It allocates model-level estimated cost across clients and token types so the resulting rows reconcile to OpenAI's imported model-level estimated cost. These client- and token-type dollar amounts are allocation estimates, not separately billed amounts reported by OpenAI.

10. Are there Active Resources for OpenAI Codex?

No. Active Resources for OpenAI Codex are not available at launch.

11. How often does Codex data refresh in the Vantage console?

Codex data refreshes daily in the Vantage console.

12. What happens if I remove the OpenAI Codex integration?

If you remove your OpenAI Codex integration, Vantage schedules background cleanup of its imports and associated costs. The data may remain visible briefly while that cleanup completes.

13. How far back will the data go?

On initial connection, Vantage attempts to import up to the most recent 90 days of Codex data available from the Analytics API.

14. Can I have multiple OpenAI Codex integrations?

Yes. You can connect multiple workspaces/organizations as separate integrations in Vantage, each with its own Analytics API key.

15. Does this integration cover my OpenAI Platform (API) spend too?

No. This connection imports Codex Analytics data for a ChatGPT workspace. Vantage's existing OpenAI Platform integration separately imports developer API costs using an OpenAI Platform Admin API key. Customers can connect OpenAI Platform, OpenAI Codex, or both; both appear under the OpenAI provider in Cost Reports, and Codex Analytics rows are identified by the openai:billing_source=codex_analytics tag. OpenAI Platform rows are identified by the openai:billing_source=admin tag.

16. Does Vantage store prompt or code content?

No. Vantage does not ingest prompts, generated code, or other user content. It ingests daily aggregated estimated cost and token usage data, which can include user email and user ID for reporting and allocation.

17. I have custom OpenAI pricing. Do you support this?

Please contact support@vantage.sh if you have negotiated pricing you would like supported.

18. Does this cover OpenAI usage routed through Azure OpenAI or other clouds?

No. This connection covers Codex usage billed through your ChatGPT workspace. OpenAI models accessed through Azure OpenAI are already captured under your Azure cost integration in Vantage, and your programmatic OpenAI Platform (developer API) spend is covered by the separate OpenAI Platform integration. The Codex Analytics API only covers Codex consumption billed directly by OpenAI for the connected workspace.