# Rush vs Codex cloud

Both run Codex in an isolated cloud machine. OpenAI's runs from ChatGPT; Rush's runs from an API with an API key and comes back with the transcript.

Facts checked 15 September 2026. Prices and limits move. Read the citations.

**The short answer.** Pick Codex cloud when you live in ChatGPT and want cloud tasks in the plan. Pick Rush when Codex should run from your own code, next to Claude Code and OpenCode.

## The useful distinction

Codex cloud: OpenAI's coding agent, running background tasks in isolated ChatGPT cloud environments. Included with ChatGPT plans. Not the 2021 Codex model.

The useful split is who builds the agent. Most products here run their own. Rush runs Claude Code, Codex or OpenCode, the vendors' own CLIs, and returns the result with the record of how it was made.

## Buyer questions

| Question | Rush | Codex cloud |
| --- | --- | --- |
| What runs | Claude Code, Codex or OpenCode, the vendors' own CLIs, or a saved agent built on one. | Codex agent. AGENTS.md for lint/test/review rules. |
| Isolation | One E2B microVM per run with its own disk, memory, desktop and browser. It parks between turns and is removed on cancel or after 7 idle days. | Isolated OpenAI-managed containers. Setup has network and secrets; agent phase is offline by default and secrets are removed first. |
| Result and pull request | A run on a repository works on its own branch and opens the pull request; a run whose checks fail ends needs_review, not completed. Reports and files come back as artifacts. | Connect GitHub or GitLab (Beta). Review the summary and diff, then open a PR. @codex on issues and PRs. |
| MCP | Rush is an MCP server at api.prix.dev/mcp with 18 tools. A run also takes MCP servers by connection or inline. | Official MCP config is documented for desktop, CLI, and IDE. Whether cloud containers load config.toml MCP servers is not documented. |
| Team | Everyone in the org sees the same Projects and runs, with who started each. Automations start runs from GitHub events, schedules, webhooks and Linear. | Parallel cloud chats. Slack and Linear. Workspace-shared environments on team plans. |
| Starting price | Free 5 agent-hours / month; Pro $20; Team $200 | ChatGPT Plus $20 / month; Pro 5x $100 / month; Free includes limited Codex |
| Where it taps out | Three harnesses. Repositories through the GitHub App only. No self-host or customer VPC. A key runs 4 at once by default, up to 100. | API-key path has no cloud, GitHub review, or Slack. Agent internet is off by default. Cloud chats can cost more than local. |

## When each side wins

### Rush wins when

- You want to start Codex runs from your own server with an API key.
- The agent needs the network while it works, or a browser and a desktop.
- Claude Code and OpenCode should run in the same Project.

### Codex cloud wins when

- You already pay for ChatGPT, and cloud tasks come with the plan.
- You want the agent offline during its work phase, with secrets removed first.
- You want @codex on GitHub issues and pull requests, and GitLab in beta.

## Where Rush sits

Rush runs the Codex CLI itself, on your ChatGPT subscription or OpenAI key, in its own microVM with network access, and returns the pull request and transcript to the API call that started it.

## Skip Rush if

- Rush runs three agents: Claude Code, Codex and OpenCode. It does not host an agent you wrote yourself.
- Repositories come through the Rush GitHub App. A GitLab or Bitbucket repository cannot be a run's repository yet.
- There is no self-hosted or customer-VPC deployment, and we do not publish a compliance report today.
- Compute is metered. Free includes 5 agent-hours a month and Pro 40, each with a weekly ceiling.
- You review the pull request. Rush does not merge for you.

## Questions people actually ask

### Can I use Codex cloud with an API key?

No. OpenAI's Codex pricing docs retrieved 15 September 2026 say the API-key path has no cloud tasks, no GitHub code review and no Slack. Rush runs Codex with an API key or a ChatGPT subscription.

### Can I move from Codex cloud to Rush?

Yes. The steps, from OpenAI's own docs, are on the Migrate from Codex cloud to Rush page. Code you already pushed stays where it is.

### Does Rush need my own model account?

By default a run uses your own Claude or ChatGPT subscription or API key, and your provider bills the model calls. Rush meters the sandbox compute against your plan, and parked time is free.

## Keep reading

- [Codex cloud alternatives](https://getrush.ai/compare/codex-cloud-alternatives.md)
- [Migrate from Codex cloud to Rush](https://getrush.ai/compare/migrate-from-codex-cloud.md)
- [Devin vs Codex cloud](https://getrush.ai/compare/devin-vs-codex-cloud.md)
- [Claude Code on the web vs Codex cloud](https://getrush.ai/compare/claude-code-web-vs-codex-cloud.md)
- [Codex cloud vs Google Jules](https://getrush.ai/compare/codex-cloud-vs-jules.md)

## Sources

1. Rush, "Rush product requirements", retrieved 2026-09-15. https://getrush.ai/cloud
2. Rush, "Rush pricing (live contract on getrush.ai)", retrieved 2026-09-15. https://getrush.ai/pricing
3. Rush, "Isolation, plainly", retrieved 2026-10-17. https://getrush.ai/docs/concepts/isolation
4. Rush, "Results and proof", retrieved 2026-10-17. https://getrush.ai/docs/concepts/results-and-proof
5. Rush, "Usage and billing", retrieved 2026-10-17. https://getrush.ai/docs/concepts/usage-and-billing
6. Rush, "Rush MCP server", retrieved 2026-10-17. https://getrush.ai/docs/concepts/mcp-server
7. Rush, "Automations", retrieved 2026-10-17. https://getrush.ai/docs/concepts/automations
8. OpenAI / ChatGPT Learn, "Codex cloud", retrieved 2026-09-15. https://learn.chatgpt.com/docs/cloud
9. OpenAI / ChatGPT Learn, "Cloud environments", retrieved 2026-09-15. https://learn.chatgpt.com/docs/environments/cloud-environment.md
10. OpenAI, "Pricing | ChatGPT Learn", retrieved 2026-09-15. https://learn.chatgpt.com/docs/pricing
11. TechCrunch, "OpenAI launches Codex, an AI coding agent, in ChatGPT", published 2025-05-16. https://techcrunch.com/2025/05/16/openai-launches-codex-an-ai-coding-agent-in-chatgpt/