OpenAI Plans to End Direct Cursor Model Access on November 12
OpenAI says it intends to wind down the contract that gives Cursor direct access to OpenAI models after Cursor's acquisition by SpaceX. The proposed shutoff date is November 12, 2026, but OpenAI says the final termination date is not yet confirmed. During the transition, users can consider an OpenAI API key for supported local Chat and Agent requests, the Codex IDE extension, or a compatible gateway; those options do not cover Cursor Tab, Auto, Cloud Agents, Background Agents, Automations, the CLI, API, or SDK.
OpenAI says it intends to wind down its contract providing OpenAI models directly to Cursor after Cursor’s acquisition by SpaceX. The company has proposed November 12, 2026 as the date when Cursor’s direct access to those models would end, but OpenAI’s user guidance says the official termination date is not yet confirmed between the two companies. (OpenAI’s announcement; OpenAI Help Center)
This is a meaningful change for developers who use GPT models through Cursor’s own model picker. It is not an immediate shutdown: OpenAI says it is giving the maximum notice allowed by its contract, and its Help Center says the current models would continue during the proposed transition period. Cursor says it is speaking with OpenAI about continuity and will share more around access. (OpenAI’s announcement; Cursor community announcement)
Key takeaways
- The proposed date is November 12, 2026. Treat it as a proposed transition date for now, not a confirmed final termination date.
- The affected path is direct Cursor access to OpenAI models. OpenAI says it will not provide future models to Cursor under the contract while it holds the contract to the latest date it can.
- The acquisition context matters. Cursor’s own announcement says it is now part of SpaceX. OpenAI says its decision follows that change of control and concerns about whether SpaceX will use OpenAI technology within its terms.
- Local Chat and Agent users have alternatives. OpenAI documents bring-your-own-key and compatible gateway routes for supported local requests, plus the separate Codex IDE extension.
- Those alternatives are not a universal replacement. OpenAI says API keys and gateway credentials do not apply to Cursor Tab, Auto, Cloud Agents, Background Agents, Automations, the Cursor CLI, API, or SDK.
What OpenAI is proposing
OpenAI’s announcement says it notified SpaceX that it intends to wind down the contract that provides OpenAI models to Cursor. The company describes November 12, 2026 as a proposed shutoff date, chosen to maximize the time developers can retain access. It also says it will not provide future models to Cursor under the contract while it waits for the cancellation window. (OpenAI’s announcement)
OpenAI gives the reason as a change of control and its assessment that it cannot be confident SpaceX will use the technology within OpenAI’s terms of service. The announcement cites OpenAI’s stated experience with contract violations involving Elon Musk’s companies. This is OpenAI’s explanation for its decision, not a prediction about what Cursor users will ultimately be able to access. (OpenAI’s announcement)
Cursor announced on August 14 that it had officially become part of SpaceX, saying the combination would give Cursor access to SpaceX’s computing capacity and support work on Grok, Grok Build, Grok Bot, the Grok API, and Cursor. That post provides the context for OpenAI’s change-of-control announcement; it does not state an OpenAI model-access transition plan. (Cursor’s SpaceX announcement)
What is confirmed versus still open
The safest way to read the announcement is to separate the contract decision from the date and from the user experience that follows.
| Question | What the official guidance says |
|---|---|
| Is the contract wind-down planned? | Yes. OpenAI says it intends to wind down its contract providing OpenAI models to Cursor. |
| What date has OpenAI proposed? | November 12, 2026. |
| Is that the confirmed final termination date? | Not yet. OpenAI’s Help Center says the official termination date is still to be confirmed between OpenAI and Cursor. |
| Could access end earlier? | Yes. OpenAI’s Help Center says Cursor may choose to end access sooner. |
| Will current models remain available during the proposed transition? | OpenAI says it would continue providing Cursor access to the OpenAI models it uses today during that period. |
| Will Cursor receive future OpenAI models through the contract? | OpenAI says it will not provide future models to Cursor under the contract. |
That distinction matters if you are planning a migration. Do not treat November 12 as a guaranteed final date, and do not assume that a new OpenAI model will appear in Cursor’s native model picker during the notice period. (OpenAI Help Center)
Options for local Chat and Agent work
OpenAI’s Help Center documents three routes. They cover supported local work in the Cursor desktop app; they are not replacements for every Cursor surface.
1. Bring your own OpenAI API key
In Cursor, an OpenAI API key can continue to power supported local Chat and Agent requests. The API account is separate from a Cursor or ChatGPT subscription, requires billing to be configured, and is billed at OpenAI API rates. The documented setup is:
- Create an OpenAI API key with billing configured.
- Open Cursor Settings → Models.
- Paste the key into the OpenAI API key setting and verify it.
- Enable the OpenAI model you want to use.
- Select that model in Cursor Chat or Agent.
The available models and limits depend on the OpenAI API organization. OpenAI also explicitly says that a ChatGPT subscription does not include OpenAI API usage. (OpenAI Help Center)
2. Use the Codex IDE extension
The Codex IDE extension is a separate agent experience inside Cursor. OpenAI’s documented path is to install the cursor:extension/openai.chatgpt extension, open Codex from the Activity Bar, and sign in with an eligible ChatGPT subscription or choose Use API Key for an OpenAI API account. (OpenAI Help Center)
This route does not turn Cursor’s own model picker back into an OpenAI-powered service. OpenAI says the extension does not change the model used by Cursor Chat, Cursor Agent, Cursor Tab, Cursor Auto, Cursor Cloud Agents, or other Cursor services. Think of it as adding Codex alongside Cursor’s native surfaces, not as restoring the direct partnership. (OpenAI Help Center)
3. Use a compatible AI gateway
Organizations can connect supported local Cursor Chat and Agent requests to OpenAI models through a compatible gateway or hosting provider. OpenAI lists Amazon Bedrock, Azure, OpenRouter, Vercel AI Gateway, and other OpenAI-compatible gateways as examples. The exact model availability, controls, credentials, and endpoint configuration depend on the provider. (OpenAI Help Center)
The important limitations
The migration options are narrower than the phrase “keep using OpenAI models in Cursor” suggests. OpenAI says bring-your-own-key and gateway credentials do not apply to:
- Cursor Tab
- Cursor Auto
- Cursor Cloud Agents
- Cursor Background Agents
- Cursor Automations
- Cursor CLI
- Cursor API
- Cursor SDK
They apply to supported local Chat and Agent requests in the desktop app. That means a team can preserve a local Chat or Agent workflow with its own OpenAI billing, but should not assume the same key will cover autocomplete, automatic routing, hosted agents, or automation services. (OpenAI Help Center)
Cursor’s own community guidance says it will continue to offer ways to access models in Cursor and that models from Anthropic, Google, Meta, and Grok will continue through direct partnerships. It does not yet provide a final OpenAI continuity plan, so treat that statement as a commitment to keep model choice available rather than as a confirmed replacement for every OpenAI-powered Cursor feature. (Cursor community announcement)
What developers should do now
If your workflow depends on OpenAI models inside Cursor, inventory the exact surfaces you use before choosing a workaround:
- Local Chat or Agent: test an OpenAI API key or a compatible gateway, keeping separate API billing in mind.
- Codex workflows: test the Codex IDE extension and confirm whether its separate agent experience fits your repository workflow.
- Tab, Auto, hosted agents, Automations, CLI, API, or SDK: do not assume a key or gateway will transfer; wait for Cursor’s continuity details and evaluate other supported models.
- Team and enterprise controls: document who owns API credentials, model allowlists, logs, and billing before changing providers.
- Dates: track the official OpenAI and Cursor updates rather than treating November 12 as final until the companies confirm it.
The practical decision is not simply “OpenAI or no OpenAI.” It is which Cursor surfaces must remain unchanged, which can move to a separately billed API path, and whether a Codex extension or another model provider is acceptable for the rest.
Bottom line
OpenAI is proposing to end its direct Cursor partnership after Cursor became part of SpaceX, with November 12, 2026 as the proposed shutoff date. The final termination date is not confirmed, and OpenAI says current model access would continue during the proposed transition period. For local Cursor Chat and Agent work, OpenAI documents an API-key route, compatible gateways, and the separate Codex IDE extension; none of those options currently covers Cursor’s hosted or automation surfaces. (OpenAI’s announcement; OpenAI Help Center)
Sources
- OpenAI, “Our decision on Cursor following its acquisition by SpaceX,” August 29, 2026: https://openai.com/index/our-decision-on-cursor-following-its-acquisition-by-spacex/
- OpenAI Help Center, “Using OpenAI models in Cursor,” checked August 29, 2026: https://help.openai.com/en/articles/20001506-using-openai-models-in-cursor
- Cursor, “Cursor is now a part of SpaceX”: https://cursor.com/blog/joining-spacex
- Cursor Community, “I heard OpenAI is about to cut off GPT access for Cursor,” checked August 29, 2026: https://forum.cursor.com/t/i-heard-openai-is-about-to-cut-off-gpt-access-for-cursor-is-this-true/169876
Read next
Keep building the workspace playbookCursor Origin Enters Early Beta: Code Hosting, Pull Requests, and GitHub Sync
Cursor's Origin code-hosting platform is now rolling out in early beta to paid plans, with hosted repositories, pull requests, code browsing, GitHub sync, and agents in every repo.
Codex CLI 0.130.0 Adds `remote-control`, Richer Plugin Sharing Metadata, and Better App-Server Thread Paging
OpenAI shipped Codex CLI 0.130.0 in May 2026. The release adds a new `codex remote-control` command for starting a headless, remotely controllable app-server, improves app-server clients with paging options for large threads (unloaded/summary/full turn items), expands plugin sharing with link metadata and discoverability controls, adds Bedrock auth support for AWS console-login credentials from `aws login` profiles, and fixes several app-server/thread reliability issues.
ChatGPT Voice Can Start and Steer Codex Tasks on Desktop, With Mobile Remote Access
ChatGPT Voice now provides a hands-free way to start, monitor, and steer Codex work. OpenAI says Voice can coordinate Codex tasks in the ChatGPT desktop app, while mobile access is available through Remote on iOS after pairing an iPhone with a desktop host. Voice conversations use the Codex task’s usage budget and require a new chat or task to begin in voice mode.
Frequently Asked Questions
Is OpenAI cutting off Cursor immediately?
No. OpenAI says it is giving notice and has proposed November 12, 2026 as the shutoff date. Its Help Center says the official termination date is not yet confirmed between OpenAI and Cursor, and Cursor may choose to end access sooner.
What is changing for Cursor users?
OpenAI says it intends to wind down its contract providing OpenAI models directly to Cursor and will not provide future models to Cursor under that contract. The developers most affected are people who rely on OpenAI models through Cursor.
Can I keep using OpenAI models in Cursor with my own API key?
OpenAI's guidance says a user can bring an OpenAI API key for supported local Cursor Chat and Agent requests. API usage is billed separately through the OpenAI API account, and the key does not cover Cursor Tab, Auto, Cloud Agents, Background Agents, Automations, the Cursor CLI, API, or SDK.
Can I use Codex inside Cursor instead?
OpenAI says users can install the Codex IDE extension in Cursor and sign in with an eligible ChatGPT subscription or use an OpenAI API key. The extension is a separate Codex experience and does not change the models used by Cursor Chat, Agent, Tab, Auto, Cloud Agents, or other Cursor services.