Compare · Cloud agents · verified 15 Sep 2026

Devin vs Codex cloud

Devin is a product you buy separately; Codex cloud comes with ChatGPT. Both work in an isolated cloud machine and aim at a pull request.

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

Start a TaskAll comparison pages
Vendor agentDevin · Cursor · Copilot · Julessame jobRushYour harness, isolated, one PR
Cloud agent loop from request to pull request

Neither product is ours. Rush is, and it gets one paragraph below. Every cell is from the vendor's own pages.

The short answer

Pick Devin for Cognition's agent with deep integrations. Pick Codex cloud if you already pay for ChatGPT and want OpenAI's agent in the plan.

Read this first

The useful distinction

Devin: Cognition's own autonomous software engineer. It plans, writes, tests, and ships in a cloud Devbox. The Brain always stays in Cognition's cloud.

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

Head to head

Buyer questions

QuestionDevinCodex cloud
What runsCognition's compound agent. Third-party LLM API keys are not supported.Codex agent. AGENTS.md for lint/test/review rules.
IsolationOne isolated VM (Devbox) per session. Snapshot boots; session changes do not write back. Outposts can run execution on customer machines; the Brain stays at Cognition.Isolated OpenAI-managed containers. Setup has network and secrets; agent phase is offline by default and secrets are removed first.
Result and pull requestCreates PRs on GitHub, GitLab, Bitbucket, or Azure DevOps. Stacked PRs on GitHub.com only.Connect GitHub or GitLab (Beta). Review the summary and diff, then open a PR. @codex on issues and PRs.
MCPstdio, SSE, and HTTP MCP client, plus an official Devin MCP server at mcp.devin.ai.Official MCP config is documented for desktop, CLI, and IDE. Whether cloud containers load config.toml MCP servers is not documented.
TeamSession UI with shell, IDE, and browser. Slack and Teams. Linear and Jira ticket triggers. Invite-team on Team and Enterprise.Parallel cloud chats. Slack and Linear. Workspace-shared environments on team plans.
Starting pricePro $20 / month; Teams $80 minimum + $40 / full seatChatGPT Plus $20 / month; Pro 5x $100 / month; Free includes limited Codex
Where it taps outYou cannot bring your own model keys. Ambiguous senior-scope work is out of scope by Cognition's own 2025 review. Quotas are opaque.API-key path has no cloud, GitHub review, or Slack. Agent internet is off by default. Cloud chats can cost more than local.

Decision

When each side wins

Devin wins when

  • You want Cognition's agent with Slack, Teams, Linear and Jira triggers built in.
  • You need pull requests on GitHub, GitLab, Bitbucket or Azure DevOps, and stacked pull requests on GitHub.
  • You need Enterprise dedicated VPC or Outposts that run execution on your machines.

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.

Rush

Where Rush sits

Rush runs Codex too, from an API with an API key, which the API-key path of Codex cloud does not offer.

Honest limits

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.

FAQ

Questions people actually ask

Is Devin cheaper than Codex cloud?

List prices, read off each vendor's pricing page on 15 September 2026. Devin: Pro $20 / month; Teams $80 minimum + $40 / full seat. Codex cloud: ChatGPT Plus $20 / month; Pro 5x $100 / month; Free includes limited Codex. The units differ, so price the job you run, not the plan.

Where does Rush fit next to Devin and Codex cloud?

Rush runs Codex too, from an API with an API key, which the API-key path of Codex cloud does not offer.

Cluster

Keep reading

Citations

Sources

Every factual claim on this page traces to one of these URLs. Retrieval date is 15 September 2026 unless the source prints an earlier published date.

  1. [1] Cognition, “Devin | The AI Software Engineer”, retrieved 2026-09-15. https://devin.ai/
  2. [2] Cognition, “Plans and Pricing | Devin”, retrieved 2026-09-15. https://devin.ai/pricing
  3. [3] Cognition Docs, “Introducing Devin”, retrieved 2026-09-15. https://docs.devin.ai/get-started/devin-intro
  4. [4] Cognition Docs, “Self-serve plans”, retrieved 2026-09-15. https://docs.devin.ai/admin/billing/self-serve
  5. [5] Cognition Docs, “Enterprise Deployment”, retrieved 2026-09-15. https://docs.devin.ai/enterprise/deployment/overview
  6. [6] Cognition Docs, “Devin GitHub integration”, retrieved 2026-09-15. https://docs.devin.ai/integrations/gh
  7. [7] Cognition Docs, “Devin MCP servers and marketplace”, retrieved 2026-09-15. https://docs.devin.ai/work-with-devin/mcp
  8. [8] Cognition, “New self-serve plans for Devin”, published 2026-04-14. https://cognition.com/blog/new-self-serve-plans-for-devin
  9. [9] Cognition, “Privacy Policy”, published 2026-03-09. https://cognition.com/legal/privacy-policy
  10. [10] OpenAI / ChatGPT Learn, “Codex cloud”, retrieved 2026-09-15. https://learn.chatgpt.com/docs/cloud
  11. [11] OpenAI / ChatGPT Learn, “Cloud environments”, retrieved 2026-09-15. https://learn.chatgpt.com/docs/environments/cloud-environment.md
  12. [12] OpenAI, “Pricing | ChatGPT Learn”, retrieved 2026-09-15. https://learn.chatgpt.com/docs/pricing
  13. [13] 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/
  14. [14] Rush, “Rush product requirements”, retrieved 2026-09-15. https://getrush.ai/cloud
  15. [15] Rush, “Rush pricing (live contract on getrush.ai)”, retrieved 2026-09-15. https://getrush.ai/pricing
  16. [16] Rush, “Isolation, plainly”, retrieved 2026-10-17. https://getrush.ai/docs/concepts/isolation
  17. [17] Rush, “Results and proof”, retrieved 2026-10-17. https://getrush.ai/docs/concepts/results-and-proof
  18. [18] Rush, “Usage and billing”, retrieved 2026-10-17. https://getrush.ai/docs/concepts/usage-and-billing
  19. [19] Rush, “Rush MCP server”, retrieved 2026-10-17. https://getrush.ai/docs/concepts/mcp-server
  20. [20] Rush, “Automations”, retrieved 2026-10-17. https://getrush.ai/docs/concepts/automations

Next

Run the next Task on Rush

Connect a repository. Pick Claude Code, Codex, or OpenCode. Review the PR.

How the runtime works