“We chose Codex because it can work directly across a large, multi-repository codebase and carry implementation through testing and verification. It helps our small team move faster on real engineering work.”
AI Coding AgentsMaker says so · source ↗
OpenAI Codex
OpenAI's coding agent, available as an open-source terminal CLI, an IDE extension, a desktop app and cloud tasks that run in the background and review pull requests.
145 makers' products we track use OpenAI Codex. For AI coding agents it's 2nd most chosen of 6 options: 145 of the 909 makers' products we know the AI coding agents of (16%). Among the 657 launched in 2026, 19% chose it, against 8% of those launched in 2024 or earlier. What makers like most: Takes complete repo-wide tasks, from implementation through tests, debugging and release prep, and verifies them in the same environment. On GitHub, 76 public pull requests moved projects to OpenAI Codex from tools we list, and 85 moved them away.
Profile checked
Works with AI agents:llms.txt- Oct 2026 OpenAI Codex got 15 new Product Hunt reviews (now 111). source
How makers use OpenAI Codex
Makers' products we track use OpenAI Codex for AI coding agents (145), most often alongside Vercel (28%), Next.js (26%), Resend (19%).
What OpenAI Codex is used for
OpenAI Codex vs the alternatives makers chose
For AI coding agents, 145 of the 909 makers' products we know the choice of picked OpenAI Codex (16%), behind Claude Code (85%). Among those launched in 2026 it's 19%.
| Option | Makers' products | 2026 launches | Free for | Team of 5, 10,000 users | Open source | |
|---|---|---|---|---|---|---|
| 772 85% | 84% | None | $100/mo · Pro | No | Compare → | |
| 145 16% | 19% | None | $40/mo · ChatGPT Go | Yes | ||
| 44 5% | 5% | None | $50/mo · Go | Yes | Compare → | |
| 27 3% | 2% | None | $50/mo · Pro | No | Compare → | |
| 4 0% | 0% | — | — | Yes | ||
| 1 0% | 0% | — | — | Yes |
For AI coding agents: how many of the 909 makers' products we know the AI coding agents of chose each option, and the share among the 657 launched in 2026, since a tool that came first has products from every year, and what each costs for a team of 5 with 10,000 users.
Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.
Often paired with
Share of the 145 makers' products on OpenAI Codex that show this choice.When to use OpenAI Codex, and when not
Pick OpenAI Codex if
- You're comfortable in a terminal and want to hand off whole tasks. AI Coding Agents →
- You already pay for ChatGPT. AI Coding Agents →
- You already use Claude Code or Codex in the terminal and want it in your editor too. AI Code Editors →
Something else fits better if
- You want to switch between model providers freely or keep your tools open source: OpenCode or Pi or goose.
- You want to try a terminal agent without paying for a subscription: Gemini CLI.
- Tasks start as GitHub issues and you want pull requests back without running anything locally: GitHub Copilot or Jules.
- Tasks arrive in Slack, Linear or a ticket tracker and you want an agent to pick them up there: Devin or Factory.
- You want agents running on a machine you control, with any model including local ones: OpenHands.
Codex is OpenAI's coding agent. The same agent works locally, in the open-source Codex CLI, the IDE extension and the ChatGPT desktop app, where it inspects a repository, edits files and runs your tools. It also works in Codex Cloud, where each task gets its own workspace and comes back as changes you can commit or open as a pull request.
More about OpenAI Codex
Locally, an OS-enforced sandbox limits writes to the current workspace and keeps network access off by default, and an approval policy decides when the agent must stop and ask. Project instructions go in AGENTS.md, and skills, plugins, MCP servers and subagents extend it. codex exec runs it non-interactively in scripts and CI. The IDE extension works in VS Code, Cursor and Windsurf, while Xcode and JetBrains IDEs offer Codex through their own integrations.
You sign in with a ChatGPT account, which uses your plan's allowance, or with an API key billed at API rates. Cloud environments are set up from GitHub repositories: Codex installs dependencies, tests the setup, and lets you allow specific network destinations and secrets, and tasks keep running while your computer sleeps. The CLI can also run against local open-source models through Ollama or LM Studio.
Codex Cloud requires ChatGPT sign-in, so with an API key alone you are limited to local work and some features are unavailable. Data handling follows whichever account you sign in with, ChatGPT workspace settings or API organization settings.
How OpenAI Codex itself is built
2 tools, from its own code, website and Product Hunt page.
OpenAI Codex pros and cons
Themes that recur in makers' words and Hacker News comments, each linked to what it summarises, with how Product Hunt tags its reviews.
Pros
- Takes complete repo-wide tasks, from implementation through tests, debugging and release prep, and verifies them in the same environment. PHPH 2PH 3PH 4
- Particularly strong at backend architecture and tracking down bugs from vague descriptions. PHHN
- Usage limits are generous compared with Claude Code and Cursor. PHPH 2
- The Rust CLI is compact and snappy, with local sandboxing, approvals and scriptable runs. PHPH 2PH 3
What Hacker News says about OpenAI Codex
On Hacker News, 21% of comments about OpenAI Codex recommend it and 12% criticize it: net praise +9%, 6th best liked of 9 tools HN discusses for AI coding agents. Commenters chose it 4.3× as often per month in 2026 as in 2025.
By year
| Year | Comments choosing it, per month | Net praise |
|---|---|---|
| 2023 | 0.6 | +6% |
| 2024 | 0.0 | — |
| 2025 | 19 | +12% |
| 2026 (8 mo) | 83 | +8% |
In their words
- “Their ACP integration is good (finally), at least with Codex.” HN, Aug 2026
- “these days I use Codex to great effect to enhance my emacs setup” HN, Jun 2026
- “every time I use Codex unless I specifically give it the abstractions it writes code that is way too specific” HN, Jul 2026
Hacker News comments January 2023 to August 2026 that name OpenAI Codex in a sentence like "I use" or "we moved to", Claude read a sample of 77 and the counts scale it to all of them; net praise is comments recommending it minus comments criticizing it, as a share of all comments about it. How it's counted, and how HN compares with makers.
Reliability and open issues
OpenAI Codex's status page posted 5+ incidents in the last 90 days, none major; the most wanted open issue on GitHub has 501 👍.
Incidents on its status page
5+ in the last 90 days · 5+ in the last year- minor Degraded performance in Codex and ChatGPT Work Oct 2026
- minor Degraded performance in GPT-6.1 Sol Codex and Work mode Oct 2026
- minor Degraded performance in Codex and Work mode: Dot turn failures, delayed responses, and reduced proactivity Oct 2026
Most wanted on GitHub
- LSP integration (auto-detect + auto-install) for Codex CLI👍 501 · opened Jan 2026 · active Sep 2026
- Please make "/undo" back👍 467 · opened Jan 2026 · active Oct 2026
- Codex Desktop for macOS repeatedly triggers `syspolicyd` / `trustd` CPU and memory runaway👍 398 · opened Jun 2026 · active Sep 2026
- IDE-integrated diff / approval👍 235 · opened Aug 2025 · active Sep 2026
- Voice transcription for IDE extension👍 236 · opened Aug 2025 · active Sep 2026
Who switches
More projects move away from OpenAI Codex: 76 pull requests in, 85 out; the most come from Claude Code and leave for Claude Code.
Claude Code → OpenAI Codex34+ PRs
- Migrate repository guidance from Claude Code to Codexyuru-sha/fuji-calendar · 2026-10-02
- chore: migrate Claude Code workflow to CodexJoycai/Jellyfin-Media-Management-Tool · 2026-10-01
- feat: add live Claude Code to Codex context handoff pathavaargsh/agent-control-plane · 2026-10-01
- Migrate project agent workflow from Claude Code to CodexJoycai/Joycai-Image-Ai-Toolkits · 2026-10-01
- Add post on switching from Claude Code to Codexnninnis/nninnis.github.io · 2026-09-22
- Confine Claude Code runs to the grants Codex getsdark-factory-build/dark-factory · 2026-09-20
- Prototype Claude Code delegation to metered native Codex workersgremlord/gremlord · 2026-09-17
- chore: migrate Claude Code skills to Codexbentleypark/aiwatch · 2026-09-04
- Change first slide title from Claude Code to Codexnarze/claude-code-in-the-cloud-slides · 2026-08-26
- Migrate Claude Code configuration to Codexrjdellecese/confect · 2026-08-24
Cursor → OpenAI Codex16 PRs
- fix(harness): pin cursor and codex lane reviewers to Codex Sol/Terrathinmansoftware/bdc-harness · 2026-09-24
- Drop Cursor ACP so Bernise talks only to Codexjoshborseth/bernise · 2026-08-29
- feat: drop cursor target, sync skills to codex explicitlychris-nowicki/ai-dotfiles · 2026-08-28
- fix(install): --update re-detects opencode/hermes/cursor instead of falling back to codexfujibee/agmsg · 2026-08-18
- chore(codex): attach .cursor rules to Codex sessions via lifecycle hookscoddy-project/coddy-agent · 2026-08-02
- Migrate Single-Agent executor from Cursor to Codexnotekmitl/knowme · 2026-07-28
- chore: rename .cursor to .codex for Codex sprintTroy-LL/cody · 2026-07-18
- Mirror Cursor skills and AGENTS.md to Claude and Codexabegong/katalyst · 2026-06-20
- [codex] Add Cursor Agent to Codex workspaceboxp/arch · 2026-06-14
- docs(agents): migrate Cursor rules to CodexDeadwood-ai/deadtrees · 2026-05-05
GitHub Copilot → OpenAI Codex14 PRs
- feat: pass Copilot auth through to Codexjohn-noble-joby/cork · 2026-09-17
- feat(copilot): promote GPT-6 Astra to Codex advanced-tier defaultSignificant-Gravitas/AutoGPT · 2026-09-15
- Migrate 30 copilot workflows to codex engine + copilot/mai-code-1-flash-pickergithub/gh-aw · 2026-08-23
- Relay Copilot native web search to the Codex clienthooyao/copilot-bridge · 2026-07-20
- chore: migrate external PR review from Copilot to Codexsend/lexime · 2026-07-02
- Codex: surface Copilot's cached_tokens to Codex (was reported as 0% cache)hooyao/copilot-bridge · 2026-06-29
- [codex] Convert Copilot instructions to Codex skillsNosto/search-templates-starter · 2026-05-06
- Switch sonar-fix from Copilot to Codexjon-lipsky-sonarsource/clipper2-java · 2026-05-05
- chore(agents): convert Copilot instructions to Codex skills, remove Copilot instructions file, and map `.claude` to `.agents`Nosto/web-components · 2026-04-15
- Update Copilot CLI to 0.0.409 and Codex to 0.101.0github/gh-aw · 2026-02-13
OpenCode → OpenAI Codex12 PRs
- Bring OpenCode chats up to Codex parityMaxteabag/clarp · 2026-09-28
- feat(image): add opencode as a second harness next to codex (PE-17)adminLynere/paperclip · 2026-09-24
- Match OpenCode conversations to the Codex layoutsyeddhasnainn/sparkles-clone · 2026-09-08
- Attribute OpenCode Codex OAuth Usage to the Codex Cardmstallone/runway · 2026-09-05
- chore: migrate OpenCode commands to Codex skillsclov3r-cc/infra · 2026-08-29
- fix(opencode): promote bare Codex alias launches to Codex backend (#2066)Twigpine/openclaude · 2026-08-24
- Migration from Opencode to CodexTractatuz/ue-agent-toolkit · 2026-08-07
- feat: migrate OpenCode workflows to Codext-chov/hearthwand · 2026-07-19
- Fix wire upstreams for the OpenAI family; opencode ChatGPT OAuth routes to the Codex backendboundedhq/beagle · 2026-07-13
- fix: opencode command defaults to codex; wrap /ls entry names in back…Alice-space/alice · 2026-04-27
OpenAI Codex → Claude Code43+ PRs
- docs(blog): enlarge pro plan reopening meme in Codex to Claude Code posthon454/hon454.github.io · 2026-10-01
- docs(blog): remove image captions from Codex to Claude Code posthon454/hon454.github.io · 2026-10-01
- docs: add Codex to Claude Code desktop experience posthon454/hon454.github.io · 2026-10-01
- Add OpenAI Codex plugin marketplace to Claude Code project settingsbrandohermoso47-rgb/cliente-vault-web-app · 2026-10-01
- fix(agents): hold Codex to Claude Code's parity — kept to the browser, reading images, follow-ups that reach the model on both, and a Claude run that cannot silently lose its toolsimshaikot/browsentic · 2026-09-27
- Port Codex Mobile for Termux to Claude Codehoangkien1703/claude-mobile-termux · 2026-09-25
- Rename to WoW AI and add Codex and Grok next to Claude Codechelinho139/wow-ai · 2026-09-24
- feat: Codex as a second agent next to Claude CodeViperwow/erindi · 2026-09-24
- Switch from Codex to Claude Code and define engineer development goalsyabukichi-lab/incident-ai-assistant · 2026-09-13
- Name the Codex marketplace b-open-io to match Claude Code [OPL-4679]b-open-io/claude-plugins · 2026-09-11
OpenAI Codex → GitHub Copilot22 PRs
- Guide Codex workflows to compatible Copilot modelsgithub/gh-aw · 2026-09-05
- fix: switch AI Moderator engine from codex to copilotgithub/gh-aw · 2026-07-23
- Switch AI Moderator engine from codex to copilotgithub/gh-aw · 2026-07-20
- fix: switch AI Moderator engine from codex to copilotgithub/gh-aw · 2026-07-08
- [codex] reduce trusted PR reviewers to Cubic and Copilotwdhunter465/next-starter-template · 2026-07-02
- fix: switch daily-model-resolution engine from codex to copilot (gpt-5.4-mini invalid for Codex)github/gh-aw · 2026-07-01
- [codex] Add Recall meeting bot integration to Meeting Copilotrennanleitao/nucleus2brain · 2026-06-24
- [codex] Send internal token to setup copilotsunbeatmusicalform/sunbeat-frontend · 2026-06-19
- fix: switch daily-fact workflow engine from codex to copilotgithub/gh-aw · 2026-05-01
- fix: drop Codex CLI namespace tools before sending to CopilotMadSkittles/Router-Maestro · 2026-04-20
OpenAI Codex → Cursor12 PRs
- Codex to Cursor CLI orchestration in portable model 1.2.0MatveyTomilov/agent-operating-model · 2026-09-23
- fix(codex): disable decorative effects to prevent cursor flickerhakula139/nixos-config · 2026-09-13
- fix: clamp Codex call_id to 64 so Cursor hops resumeskillsynchq/txcript · 2026-09-11
- Add Codex Remote to Cursor ACP bridge POCkeepprogress/grammar-watch · 2026-09-01
- Relay Codex review comments to Cursor Automationsicho648/flowcrafter · 2026-08-02
- [codex] Port universal applyTo to Cursor alwaysApplymicrosoft/apm · 2026-07-05
- [codex] Anchor command runner to live cursorBarMal/Serenity · 2026-07-04
- [codex] Place command runner adjacent to cursorBarMal/Serenity · 2026-07-04
- [codex] move full CI to nightly Cursor repairImagine-That-Ai/BurnBar · 2026-06-13
- Add Codex to Cursor handoffcxk280/ship · 2026-05-28
OpenAI Codex → OpenCode8 PRs
- feat!: port /limits and Codex Usage Snapshot to OpenCode v2martinvidovic/opencode-limits · 2026-09-26
- feat(herdr): migrate team agents from codex to opencodeMrBrooks89/fedora-niri-dotfiles · 2026-09-02
- Convert Codex plugin to an OpenCode pluginTheRealDinghyDog/opencode-plugin-cc · 2026-07-08
- [codex] Restrict DuckPR review to opencodesuperduck-ai/open-managed-agents · 2026-07-07
- Revert "[codex] Migrate agent chat flow to OpenCode V2"oomol-lab/wanta · 2026-06-30
- [codex] Migrate agent chat flow to OpenCode V2oomol-lab/wanta · 2026-06-30
- feat(codex): plan-007 — new codex plugin with full v0.5.1 parity to opencodemongmong/claudecode-buddy · 2026-05-17
- refactor: migrate Codex CLI to OpenCode CLIhidemaro-nsketch/claude-code-orchestra · 2026-03-15
Public pull requests on GitHub since Oct 2024 whose title says "X to Y" — real code changes moving a project from one tool to another, by developers in general.
Apps built with OpenAI Codex
145 makers' products, each linked to the source that shows it. The best known on Product Hunt: MindPal (3,252 followers), Topview (2,655 followers), Velo (2,622 followers), Notte (1,965 followers), Fuzzy AI (1,691 followers).
“Shoutout to Codex for being part of our engineering team.”
AI Coding AgentsMaker says so · source ↗“Codex CLI is insane.”
AI Coding AgentsMaker says so · source ↗“One of the AI tools power our code base”
AI Coding AgentsMaker says so · source ↗“Our agent uses it to build the automations”
AI Coding AgentsMaker says so · source ↗“Creating beautiful interfaces and long running coding sessions”
AI Coding AgentsMaker says so · source ↗“Codex is one of the three harnesses HarnessRouter routes to, and we use it in our own build loop. Its structured output made it one of the cleanest harnesses to normalize against.”
AI Coding AgentsMaker says so · source ↗“Codex is one of the engines behind MausBot’s coding bots. Its local CLI lets us bring code-oriented agent work into a familiar chat interface while users keep their own provider account and can run Codex alongside other agents.”
AI Coding AgentsMaker says so · source ↗“Making it possible to integrate MCPs seamlessly”
AI Coding AgentsMaker says so · source ↗“Codex gives Almanac the agentic reasoning to understand company knowledge and keep it useful.”
AI Coding AgentsMaker says so · source ↗“Thanks to Tibo for repeatedly refreshing our usage quota, allowing us to consistently use GPT-5.6 Ultra High for coding and development.”
AI Coding AgentsMaker says so · source ↗“Our token compression technology has to be compatible and efficient with Codex, so we have to use it extensively to understand how we can optimize its context.”
AI Coding AgentsMaker says so · source ↗“Codex is a supported local agent in Oasis! Awesome CLI experience.”
AI Coding AgentsMaker says so · source ↗“Computer use helped me automate my testing lifecycle and act as an adversary”
AI Coding AgentsMaker says so · source ↗“Great for coding and building products like real quick.”
AI Coding AgentsMaker says so · source ↗“Codex works with me across the Chronock monorepo, from implementation and tests to refactoring and release work. Compared with editor-first tools such as Cursor or Windsurf, its agentic workflow is especially useful for completing repository-wide tasks and verifying them in the same environment. It accelerates AI-native development without replacing architecture, review, security, or product judgment; I remain responsible for the quality bar.”
AI Coding AgentsMaker says so · source ↗“OpenAI helped shape the way I built Git Pitcher. I used AI reasoning and coding workflows to move faster from product idea to working implementation, especially while planning the repo analysis, artifact structure, and launch flow.”
AI Coding AgentsMaker says so · source ↗“Low latency and great responses”
AI Coding AgentsMaker says so · source ↗“I just really like this CLI. Open AI is also super generous with usage, which we really needed when building Cosyra. My favorite thing about Codex is probably its tiny footprint. Since it's built in Rust, it's just far more compact and snappier than its counterparts in OpenCode and Claude Code.”
AI Coding AgentsMaker says so · source ↗“We built Deep Work Plan to be tool-agnostic, and Codex is a perfect example of why that matters: the plan is just files in the repo, so Codex reads it, runs the tasks, and passes the gates with zero changes. We lean on it for cloud, PR-first, parallel execution — hand it scoped tasks and let them run. It's proof that the methodology lives happily in any agent. One of the agents we use every day, each excellent in its own lane.”
AI Coding AgentsMaker says so · source ↗“The best general purpose superapp”
AI Coding AgentsMaker says so · source ↗“A core local coding-agent workflow we designed Vokal around. Codex makes individual engineering work faster; Vokal helps teams coordinate that work.”
AI Coding AgentsMaker says so · source ↗“Gave us comprehensive build and test superpowers.”
AI Coding AgentsMaker says so · source ↗“Helps us move faster from idea to implementation. Useful for prototyping, debugging, and shipping production-ready changes. A fast, reliable coding partner during development.”
AI Coding AgentsMaker says so · source ↗“Our devs use Codex too, it takes the boring parts like tests and debugging off their plate.”
AI Coding AgentsMaker says so · source ↗“Codex helped us not only speed up development but also automate testing. Computer use is dope!”
AI Coding AgentsMaker says so · source ↗“I chose Codex for its strong repository awareness and its ability to investigate bugs, modify code, run checks, and prepare focused pull requests.”
AI Coding AgentsMaker says so · source ↗“We use Codex through its CLI as a foundation and layer orchestration, parallelization, and multi-run verification on top to make it production-ready.”
AI Coding AgentsMaker says so · source ↗“Codex helps me do development, GTM, and admin tasks in the same chat surface”
AI Coding AgentsMaker says so · source ↗“It played a key role in the development of our official website, achieving excellent fidelity to the design mockups.”
AI Coding AgentsMaker says so · source ↗“Codex was instrumental to architecting and implementing the backend for Cronloop”
AI Coding AgentsMaker says so · source ↗“Codex became a practical development partner throughout the project. It helped me iterate on SwiftUI implementation, investigate recording and audio-routing edge cases, diagnose build issues, and turn product decisions into working features across Mac, iPhone, and Apple Watch.”
AI Coding AgentsMaker says so · source ↗“just better work experience.”
AI Coding AgentsMaker says so · source ↗“Codex accelerated development by handling repetitive implementation tasks, letting us spend more time refining product logic and user experience.”
AI Coding AgentsMaker says so · source ↗“One of our go-to coding agents”
AI Coding AgentsMaker says so · source ↗“We chose Codex so we could support developers with the tools they were already using.”
AI Coding AgentsMaker says so · source ↗“Unbelievable good app for coding (projects, in app browser, terminal, UI/UX).”
AI Coding AgentsMaker says so · source ↗“OpenAI Codex CLI brings frontier reasoning right into our terminal. Its agentic file editing and sandboxed test execution let us refactor code and debug without breaking our developer workflow.”
AI Coding AgentsMaker says so · source ↗“So powerful that nothing is impossible”
AI Coding AgentsMaker says so · source ↗“Understands architecture and backend needs much better than alternatives. Recommend using luna for most cost efficiency.”
AI Coding AgentsMaker says so · source ↗“Codex was especially useful when I needed a faster coding partner to continue the work while Claude Code was taking longer to respond. Claude Code helped me with the deeper architecture, planning, and product structure, but Codex became a strong assistant for execution. When I needed quick iterations, fixes, or implementation support, Codex helped me keep moving without losing momentum. It also worked really well for the visual generation side of the project. I used it while building the profile system and refining visual-related features. The image generation workflow worked very smoothly, and Codex helped me connect those ideas into the actual product experience.”
AI Coding AgentsMaker says so · source ↗“The other agent I keep running next to Claude Code, which is the reason the manager is tool-agnostic: status comes from reading the pane, so a new CLI is a few regex lines in a toml file.”
AI Coding AgentsMaker says so · source ↗“Vibe-coded thru and thru. Impossible to not use Codex for a project like this”
AI Coding AgentsMaker says so · source ↗“We use Codex as a second pair of eyes. It reviews pull requests, catches edge cases and runs long refactors in the background while we keep shipping. Strong at following instructions exactly, and great at reading a large codebase quickly.”
AI Coding AgentsMaker says so · source ↗“Sayonic works with Codex as a CLI agent, so people who already use it can drive it by voice.”
AI Coding AgentsMaker says so · source ↗“I chose Codex CLI because it combines strong coding performance with local execution, workspace sandboxing, approvals, and a scriptable CLI for launching fresh workers and critics.”
AI Coding AgentsMaker says so · source ↗“Better at backend architecture than Claude Code, better limits than cursor”
AI Coding AgentsMaker says so · source ↗None of these in makers' own words.
What its projects tell coding agents
What open-source projects' AGENTS.md and CLAUDE.md files say about OpenAI Codex, in our words — from the 15 files that name it:
Listed among several coding agents the shared instruction file is written for, alongside Claude Code and Copilot.4 files
Alternatives to OpenAI Codex
All alternatives by situation →Further reading
Questions makers ask about OpenAI Codex
Is OpenAI Codex good for production?
145 makers' products we track run on it. Its own status page posted 5+ incidents in the last 90 days; vendors decide what they post there, so this doesn't compare across tools. On GitHub, 76 public pull requests moved projects to OpenAI Codex from tools we list, and 85 moved them away.
What is OpenAI Codex used for?
OpenAI's coding agent, available as an open-source terminal CLI, an IDE extension, a desktop app and cloud tasks that run in the background and review pull requests. Makers' products we track use it for AI coding agents (145).
What are the alternatives to OpenAI Codex?
For AI coding agents, the other options makers' products we track chose most are Claude Code (772), OpenCode (44), GitHub Copilot (27), Pi (4), Aider (1). Open source or self-hostable among them: OpenCode, Pi, Aider. Each alternative, with the situation it fits, is on the OpenAI Codex alternatives page.
Is OpenAI Codex still popular with new products?
More than before: among the 657 makers' products launched in 2026 that made the AI coding agents choice, 19% chose OpenAI Codex, against 8% of the 62 launched in 2024 or earlier.
How much does OpenAI Codex cost per month?
For a team of 5 with 10,000 users, OpenAI Codex costs $40 a month on the cheapest plan that fits (ChatGPT Go), from its pricing page.
Do developers switch to or away from OpenAI Codex?
On GitHub, 76 public pull requests moved projects to OpenAI Codex from tools we list, most from Claude Code (34), Cursor (16), GitHub Copilot (14); 85 moved them away, most to Claude Code (43), GitHub Copilot (22), Cursor (12).
When should you use OpenAI Codex?
You're comfortable in a terminal and want to hand off whole tasks (AI coding agents). You already pay for ChatGPT (AI coding agents). You already use Claude Code or Codex in the terminal and want it in your editor too (AI code editors). Something else fits better if: you want to switch between model providers freely or keep your tools open source, pick OpenCode or Pi or goose; you want to try a terminal agent without paying for a subscription, pick Gemini CLI.
Can I use Codex without a ChatGPT subscription?
Yes, locally. The CLI, IDE extension and desktop app accept an OpenAI API key billed at standard API rates, but Codex Cloud needs a ChatGPT sign-in and some workspace features are unavailable with a key. source ↗
Which editors does the IDE extension support?
The Codex extension works in VS Code, Cursor and Windsurf. Xcode and JetBrains IDEs offer Codex through their own built-in integrations. source ↗
Can it run local models?
Yes. The CLI's --oss flag uses a local open-source model provider, either LM Studio or Ollama. source ↗
What can the agent do on my machine?
By default it runs in an OS-enforced sandbox that can write only to the active workspace and has no network access, and an approval policy controls when it must ask you first. You can loosen or tighten both in its configuration. source ↗
How is my data handled?
It depends on how you sign in. With ChatGPT, usage follows your workspace permissions and its retention and residency settings; with an API key, it follows your API organization's retention and data-sharing settings. source ↗
What happens when I hit my usage limit mid-task?
The agent can finish the turn it is working on. After that, Plus and Pro users can buy extra credits, and anyone can keep working locally with an API key. source ↗
Can I run Codex in CI?
Yes. Use codex exec with API key authentication for programmatic workflows such as CI jobs, and don't expose Codex execution in untrusted or public environments. source ↗
Is OpenAI Codex free?
Yes — there is a free tier a small product can run on; paid use starts at $8/mo. source ↗
Is OpenAI Codex open source or self-hostable?
Open source. source ↗
Can AI coding agents work with OpenAI Codex?
It serves an llms.txt docs index.
Who uses OpenAI Codex?
145 makers' products we track, each with a source. The best known of them on Product Hunt: MindPal (3,252 followers), Topview (2,655 followers), Velo (2,622 followers), Notte (1,965 followers), Fuzzy AI (1,691 followers). source ↗