- Browse model providers under Sessions → Model providers and choose the default for new chats.
- Enable features.tui_workspace to view a side conversation beside the main chat. Drag the divider to resize the panes or press leader z to zoom the focused pane.
- Attach recordings when using an audio-capable model. Codex skips confirmed rejected recordings on later turns and uses clip duration to keep more short recordings in context.
- Enable features.image_generation_animation to generate or edit animated GIFs with optional frame count and timing controls.
- Fullscreen Codex displays image previews and pets in tmux on Ghostty and Kitty with allow-passthrough all.
- Set tui.notification_method = "osc777" for terminal notifications. Approval notifications show longer command previews and distinguish terminal input from network access requests.
OpenAI Codex by OpenAI
OpenAI's coding agent, offered as an open-source terminal CLI, an IDE extension, a desktop app and a cloud agent that works on repositories. Users sign in with a ChatGPT account or use an OpenAI API key; with an API key, Codex works in the CLI, SDK and IDE extension without cloud features such as GitHub code review and Slack. On the Free and Go plans Codex runs in the desktop app (subject to rollout); the web, CLI and IDE extension come with Plus and higher.
Use it in: CLI IDE Desktop app Web app Mobile app API / SDK Works with: MCP Local models Own API key Multi-agent Headless / CI
On this page
Usage data: latest readings Nov 17 to Nov 19, 2026, last checked Nov 19, 06:26 UTC. Date of each number
What OpenAI Codex does
OpenAI Codex is a terminal agent + IDE extension + cloud agent from OpenAI, first released in April 2025. You use it through a command-line tool, IDE extensions, a desktop app, a web app, mobile apps and an SDK on macOS, Linux and Windows. Its documentation covers MCP servers, subagents or parallel agents, a non-interactive mode for scripts and CI, your own API key and local models. It can run models from OpenAI, Amazon Bedrock and any OpenAI-compatible endpoint. Paid plans start at $8 a month; there is a free tier.
Teams use it for: DevOps and CI, Bug fixing, Testing, Incident response, Onboarding. See examples.
Recent developments
- 2026-03: OpenAI introduced Codex Security; Codex reported at more than 2 million weekly active users (per Wikipedia)
Key facts
| Maker | OpenAI |
|---|---|
| Type | Terminal agent + IDE extension + cloud agent |
| First released | April 2025 |
| License | Apache-2.0 |
| Interfaces | CLI, IDE extension, Desktop, Web, Mobile, SDK |
| Platforms | macOS, Linux, Windows |
| Models | OpenAI GPT models (GPT-6 family per current docs) via ChatGPT sign-in or an OpenAI API key |
| Local models | `--oss` mode with Ollama or LM Studio (docs) |
| Bring your own API key | Yes |
| Main language | Rust |
| Open issues and PRs | 27,854 |
| Open VSX downloads | 13,739,564 (includes updates) |
| Homebrew installs | 529,265 in 30 days · 1,420,661 in a year |
Pricing
What each OpenAI Codex plan includes: usage limits →
Limit resets: OpenAI last reset Codex usage limits on Nov 12, 2026 (7d ago), as a reset to apply yourself. 63 resets since Sep 17, 2025, about every 6.8 days. All Codex resets →
Included in ChatGPT plans: Free $0, Go $8/month, Plus $20/month, Pro from $100/month ($100, $200 or $500 tiers), Business $20/user/month billed annually ($25 monthly), Enterprise and Edu by contact. API-key use is billed at API rates (no cloud features).
Source: official pricing page, checked Nov 5, 2026. Prices change often; confirm before you buy.
Quick answers
- Free to start?Yes
- Open source?Yes (Apache-2.0)
- Works with local models?Yes
- Bring your own API key?Yes
- Runs onmacOS, Linux, Windows
- Where you use itCLI, IDE extension, Desktop, Web, Mobile, SDK
- Supports MCP?Yes
- Model providersopenai, amazon, any OpenAI-compatible, local
- Subagents / multi-agent?Yes
- Runs in scripts / CI?Yes
- Feature comparisonAll agents
- Closest alternativeClaude Code · all alternatives
What's new in OpenAI Codex: November 2026 so far
November brought major improvements to the terminal interface and backend flexibility, including a new command center, enhanced model provider switching, and better support for multimodal inputs like images and audio recordings.
- Open the agent command center with codex agents --no-daemon to manage sessions and see unread results at a glance.
- Switch between cloud backends from the terminal without losing your work, and saved chats now stay accessible when changing providers.
- Preview local images while composing and attach audio recordings for models that support them.
- View a side conversation workspace alongside your main chat with the new TUI workspace feature.
4 releases this month · summarised from the official release notes · all months
How teams use OpenAI Codex
Auto-fix failing CI with the Codex GitHub Action
When your CI workflow fails, a second GitHub Actions workflow runs Codex on the failing commit, asks it to reproduce the failure and make the smallest fix, and opens a pull reques…
Every failed CI run triggers a Codex attempt on the exact failing commit, and any fix is saved as a patch artifact. (as reported by OpenAI)
How Eon unified multi-cloud incident tooling behind one MCP Gateway on AgentCore
Eon, a cloud data protection vendor, replaced per-tool credentials and local MCP servers with a single authenticated MCP gateway on Amazon Bedrock AgentCore. Developers and an aut…
As reported by AWS (guest post co-written with Eon): developer onboarding dropped from 2 days to 2 minutes.
Delegate a Linear issue to Codex and turn the result into a PR
With Codex installed in Linear, you can assign an issue to Codex or mention @Codex in a comment. It runs a cloud task on the linked repository and reports back in the issue.
Codex picks an environment and repository and starts a cloud task. (as reported by OpenAI)
How an OpenAI team built a product with Codex writing all of the code
A small OpenAI team built an internal beta product starting from an empty repository, with Codex writing the code and humans steering. The post covers how they set up the reposito…
As reported by OpenAI: 'roughly 1,500 pull requests have been opened and merged with a small team of just three engineers'.
How Datadog tested Codex code review by replaying past incidents
Datadog ran Codex against old pull requests that had contributed to incidents, to check whether AI review would have caught the risk. After the test, it rolled Codex review out wi…
As reported by OpenAI: Codex 'found more than 10 cases, or roughly 22% of the incidents' examined, where engineers said its feedback would have made …
Porting JustHTML from Python to JavaScript with Codex CLI in an afternoon
Simon Willison gave Codex CLI a handful of high-level prompts to port JustHTML to a dependency-free JavaScript library. He published the repo and a playground.
The library passes 9,200 tests from the html5lib-tests suite. (as reported by Simon Willison)
AgentGid articles about OpenAI Codex
JEV for Claude Code: A Second Opinion on Every Command
A viral post says jev-axi screens every command your Claude Code agent runs for a fraction of a cent. We read the repo, TypeSafe's docs and privacy policy: what runs locally, what is sent, how it fails, and how it compares with auto mode.
VoiceStudio: The Free, Local ElevenLabs Alternative
A viral post calls VoiceStudio a 100% free ElevenLabs replacement with 19.4K stars. We checked it against GitHub and the project's own pages: what it does, what is really free, the license catch in the default voice model, and how to plug it into Claude Code or Codex.
Best Claude Code Alternatives in 2026: Open-Source, Cheaper and Multi-Model
The strongest alternatives to Claude Code, grouped by why you are switching: price, model choice, open source or a different workflow. Live usage and prices for each.
Get started with OpenAI Codex
Project instructions file for OpenAI Codex: where it goes, format and a template →
| npm | npm install -g @openai/codex |
|---|---|
| Homebrew | brew install --cask codex |
| VS Code | code --install-extension openai.chatgpt |
These commands use the packages tracked on this page. The vendor may recommend a different installer; see the official documentation.
OpenAI Codex safety and permissions
| Out of the box | In a git repo: edits and runs commands inside the workspace sandbox and asks for anything beyond it; read-only in untrusted folders. |
|---|---|
| Sandbox | built-in OS-enforced sandbox (Seatbelt on macOS, bwrap+seccomp on Linux, native on Windows): read-only, workspace-write or danger-full-access; no network and writes limited to the workspace by default. |
| Approval modes | --ask-for-approval (on-request, never or granular) plus --sandbox <mode>; --dangerously-bypass-approvals-and-sandbox (--yolo) removes both. |
| Hooks | PreToolUse hooks can deny shell commands, patches and MCP calls (a guardrail, not a hard boundary). |
What OpenAI Codex users report right now
Open Request Users ask for the CLI's background daemon to be opt-in: since 0.157.0, opening an interactive session starts a persistent, self-updating background server that keeps running after Codex is closed.
Workaround: Start the CLI with --no-daemon to run a session without the shared background server.
Reported on GitHub · 192 👍
Open Users report Codex requests suddenly failing with '401 Unauthorized' and an 'Incorrect API key' message while signed in with ChatGPT. The issue is still open, and maintainers have closed several similar reports as duplicates of it.
Reported on GitHub · 105 👍
Open Users report that on Windows the Codex desktop app's floating pet ignores the mouse: clicks and drags pass through to the window underneath, so the pet cannot be moved.
Reported on GitHub · 42 👍
Fixed in 0.159.2 Users report that on Windows, after the Codex daemon arrived in CLI 0.157.0, console windows briefly flashed open during every request.
Reported on GitHub · 156 👍
Closed Users report that in macOS desktop app version 26.1016.51333, sending the first message in a new chat failed with a 'Cannot serialize value' error. The issue was closed as completed a day later; the thread does not name the fixing version.
Reported on GitHub · 170 👍
The most discussed problems in its GitHub repository, checked Nov 17, 2026. Written by AgentGid in its own words; statuses and fix versions come from the project's own pages. Reports are users' accounts, not AgentGid tests.
Run OpenAI Codex inside GitHub
Codex Exec Action: Run `codex exec` with a prompt. Used in 11 public projects set up for AI coding agents (AgentGid scan).
uses: openai/codex-action@<version> · keys: openai-api-key
OpenAI Codex Security: Scan code with Codex Security, enforce findings policy, and produce JSON and SARIF reports.
uses: openai/codex-security@<version>
Add the key in Settings → Secrets and variables → Actions, then add a workflow file with the line above. The model is paid through your own key; GitHub Actions minutes are free in public repositories.
Add-ons for OpenAI Codex: save tokens, add memory and tools
Listed because each tool documents support for OpenAI Codex. Savings and results depend on your workload.
OpenAI Codex alternatives
Best OpenAI Codex alternatives compared · All terminal coding agents
| # | Agent | Gid Score | Downloads 7d | 7d | 30d | Stars | Latest release | Price | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 |
Claude CodeAnthropic |
88 | 18.9M | up 15.1% | up 15.5% | 156K+136/day | 2.1.328yesterday | $20/mo | |||
| 3 |
OpenCodeAnomaly (SST team) |
80 | 3.9M | up 15.4% | up 28.7% | 228K+266/day | 2.3.22d ago | Free + $10/mo | |||
| 7 |
PiEarendil Inc. (created by Mario Zechner) |
73 | 6.4M | up 1.6% | up 36.3% | 123K+193/day | 1.5.210d ago | Free (OSS) | |||
| 15 |
Gemini CLIGoogle |
62 | 495K | up 16.9% | up 6.8% | 109K+35/day | 0.69.0yesterday | API usage or license | |||
| 20 |
ClineCline Bot Inc. |
58 | 84.3K | down 8.7% | up 14.9% | 71.8K+38/day | 4.1.2625d ago | Free + $9.99/mo | |||
| 22 |
GitHub Copilot CLIGitHub (Microsoft) |
56 | 1.7M | down 8.3% | up 8.3% | 11.4K+6/day | 1.0.1142d ago | Free + $10/mo | |||
Head to head: OpenAI Codex vs Claude Code · OpenAI Codex vs Cursor · OpenAI Codex vs OpenCode · OpenAI Codex vs Google Antigravity · OpenAI Codex vs GitHub Copilot · OpenAI Codex vs Pi · OpenAI Codex vs Devin · OpenAI Codex vs Gemini CLI · OpenAI Codex vs mini-SWE-agent
Tools in projects built with OpenAI Codex
Most used libraries and frameworks in 463 public GitHub repos set up for OpenAI Codex, from their dependency files.
- TypeScript65.2%
- ESLint42.5%
- Prettier32.8%
- Tailwind CSS32.2%
- Vitest31.5%
- Vite31.1%
- React28.3%
- Zod27.9%
- Playwright (JavaScript)23.3%
- Husky19.4%
Used far more often here than in AI-built projects overall: what sets OpenAI Codex projects apart.
- @nx/vitest3.7% · 6.2×
- @nx/devkit3.9% · 5.6×
- @semantic-release/git3.0% · 5.0×
- @semantic-release/changelog3.0% · 5.0×
- Nx Vite Plugin4.1% · 4.1×
- Nx JS Plugin6.3% · 3.9×
- Ultracite4.3% · 3.9×
- Nx Web4.3% · 3.9×
Ratings from people who use OpenAI Codex
No ratings yet. Used OpenAI Codex? Rate it and say what it is good and bad at.
OpenAI Codex news
-
Inside the Codex sandbox
OpenAI describes how Codex executes commands on machines while enforcing configurable filesystem and network restrictions across macOS, Linux, and Windows.
-
Introducing GPT-6.2 Astra and Sol
OpenAI released GPT-6.2 Astra and Sol models with improved professional outputs, coding, computer use, and cost efficiency across ChatGPT Work, Codex, and API.
-
Introducing ChatGPT Wallet
ChatGPT Wallet enables secure purchases for agents in ChatGPT Work, Codex, and dots with user approval and card protections.
-
OpenAI Codex 0.168.0 released
Browse model providers under Sessions → Model providers and choose the default for new chats.; Enable features.tui_workspace to view a side conversation beside the main chat. Drag the divider to resize the panes or press leader z to zoom the focused pane…
-
OpenAI Codex: a usage-limit reset to apply yourself
Activated a banked reset that was expected to land in everyone's accounts soon, after shipping some updates to dots.
-
Holiday Extras is rethinking what it can build with Codex
Holiday Extras uses Codex to improve apps and booking conversion.
-
OpenAI Codex on Amazon Bedrock
AWS published guidance on deploying OpenAI Codex through Amazon Bedrock for enterprises, addressing access, governance, and cost considerations.
-
OpenAI Codex 0.167.2 released
Codex now sends tool updates incrementally by default on Google Cloud Vertex AI. To opt out, set incremental_tools_legacy_gcp = false under [features] in config.toml.
What changed: recent OpenAI Codex releases
- Codex now sends tool updates incrementally by default on Google Cloud Vertex AI. To opt out, set incremental_tools_legacy_gcp = false under [features] in config.toml.
Earlier releases
- Full Changelog
- Open the agent command center with codex agents --no-daemon. The command center and session picker now mark unread results.
- Switch between connected backends from the terminal session picker and keep your drafts when switching.
- Preview local image attachments while composing on supported terminals, and open viewed local images from clickable transcript paths.
- Find API key and cloud provider options under “Sign in another way.” Enable gcp_setup_wizard to set up Google Cloud Vertex AI with Application Default Credentials or a credential file.
- Opt into macOS or Windows system proxy settings from /experimental.
- Expand patch details directly in the transcript while an approval prompt is open.
- Added keyboard-driven transcript selection with word, line, and page motions, integrated search, and configurable shortcuts.
- Rendered Markdown consistently in async questions, choices, answers, and transcript replies.
- Added GPT-6.1 Terra to the bundled, Vertex AI, and Amazon Bedrock model catalogs.
- Added codex queue --resume to resume unloaded threads after queueing a message.
- Added clearer safety-block explanations and selectable safer follow-up prompts in the TUI.
- Added codex doctor --network-only for focused configuration, connectivity, and WebSocket diagnostics.
- Added a built-in Google Cloud Vertex AI provider, configured with model_provider = "gcp".
- Added GPT-6.1 Terra to the model catalogs.
- Added a repository filter to the session picker and directory grouping to the agent command center, making it easier to navigate sessions across linked worktrees.
- Added experimental attached thread forks to the app-server API.
- Enabled reasoning effort changes during a conversation by default for supported models.
- Enabled memory v2 by default for fresh profiles and added support for importing external-agent memories into it.
OpenAI Codex prices over time
OpenAI Codex usage limit resets over time
OpenAI sometimes resets the 5-hour and weekly limits of Codex early, after a bug, an outage or a launch, and announces it on X.
| Last reset | Nov 12, 2026 (7d ago), to apply yourself |
|---|---|
| Average gap | 6.8 days, longest 67.7 |
| This wait | longer than 76% of past gaps |
| Resets so far | 63 since Sep 17, 2025 |
Activated a banked reset that was expected to land in everyone's accounts soon, after shipping some updates to dots. Post on X
OpenAI Codex popularity and usage data
OpenAI Codex was downloaded 26.6M times from npm in the week ending Nov 17, 2026, up 2.4% on the week before and up 7.4% over the last 30 days compared with the 30 before. Its VS Code extension has 16.9M installs and average 2.7 out of 5 across 975 marketplace ratings. The GitHub repository has 136K stars, 6,347 of them added in the last 30 days. The latest stable release is 0.168.0, published Nov 13, 2026; there were 7 stable releases in the past 30 days. On public GitHub, 277,056 pull requests were opened from branches with OpenAI Codex's default codex/ prefix in the week of Nov 9, 2026, and 87% of the pull requests from the four weeks before that were merged. On Terminal-Bench 4.1 its best listed result is 62.1% with GPT-6.2 Astra, ranking 5 of 7 agents on that board. It was mentioned in 1,429 Hacker News posts and comments over the last 30 days.
Download the raw daily series: openai-codex.csv
Gid Score breakdown
Gid Score shows public popularity and activity (usage, GitHub stars, mentions, growth), not the quality of the product. Each component is scored 0 to 100 from public signals; a dash means no data for it. How the Gid Score works.
How fresh these numbers are
| Number | Latest reading | Last checked |
|---|---|---|
| Package downloads | Nov 17, 2026 | Nov 19, 06:22 UTC |
| GitHub stars | Nov 19, 2026 | Nov 19, 06:26 UTC |
| VS Code installs | Nov 19, 2026 | Nov 19, 06:26 UTC |
| Open VSX downloads | Nov 19, 2026 | Nov 19, 06:26 UTC |
| Homebrew installs | Nov 19, 2026 | Nov 19, 06:26 UTC |
| Wikipedia views | Nov 18, 2026 | Nov 19, 06:26 UTC |
| Hacker News mentions | Nov 18, 2026 | Nov 19, 06:26 UTC |
We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.
OpenAI Codex benchmark results
| Benchmark | Model | Effort | Tasks solved | Cost of run | Place |
|---|---|---|---|---|---|
| Terminal-Bench 4.1 | GPT-6.2 Astra | max | 62.1% ± 2.3 | $1,409 | #5 of 7 |
| GPT-6.2 Sol | max | 61.5% ± 2.8 | $349 | ||
| GPT-6.1 Sol | max | 60.9% ± 3.0 | $634 | ||
| GPT-6 Astra | max | 60.3% ± 2.8 | $3,267 | ||
| Terminal-Bench 4.0 | GPT-6 Astra | max | 58.2% ± 2.8 | $3,267 | #3 of 6 |
| GPT-6.1 Sol | max | 58.2% ± 3.1 | $634 | ||
| GPT-6 Astra | xhigh | 57.9% ± 2.7 | $2,351 | ||
| GPT-6 Astra | high | 57.9% ± 3.0 | $2,269 | ||
| Terminal-Bench 3.0 | GPT-5.6 Sol | max | 34.6% ± 3.1 | $3,951 | #2 of 6 |
| GPT-5.6 Terra | max | 20.8% ± 2.7 | $2,485 | ||
| GPT-5.6 Luna | max | 14.3% ± 2.5 | $1,598 | ||
| Terminal-Bench 2.1 | GPT-6 Astra | high | 87.4% ± 1.8 | $774 | #1 of 6 |
| GPT-6 Astra | medium | 87.0% ± 1.9 | $619 | ||
| GPT-6 Astra | low | 86.7% ± 1.8 | $463 | ||
| GPT-6 Astra | max | 86.7% ± 1.4 | $1,070 | ||
| Terminal-Bench 2.0 | GPT-5.5 · v0.121.0 | — | 82.2% ± 2.2 | — | #4 of 43 |
| GPT-5.2 · v0.73.0 | — | 62.9% ± 3.0 | — | ||
| GPT-5.1-Codex-Max · v0.63.0 | — | 60.5% ± 2.7 | — | ||
| GPT-5 · v0.53.0 | — | 49.6% ± 2.9 | — |
Up to four best runs per board. Scores are not comparable between Terminal-Bench versions. Source: Terminal-Bench (tbench.ai).
Pull requests by OpenAI Codex on GitHub
In the week of Nov 9, 2026, 277,056 pull requests matching OpenAI Codex's signature (head:codex/) were opened in public repositories; 87% of those opened in the four weeks before were merged. A branch prefix identifies the tool; the pull request itself is usually opened under the developer's own account.
Reliability
From the vendor's public status page · now: All systems operational
1 incidents reported in the last 30 days, 1.0 hours in total from report to resolution.
Each bar is one month; the last one is the month so far. Vendors post even short slowdowns, so read this as how often something was reported, not as downtime. Compare with other agents →
| Oct 21, 2026, 02:26 UTC | Degraded performance in ChatGPT, Codex, and APIs · minor |
|---|---|
| Oct 14, 2026, 07:05 UTC | Degraded performance in Codex and ChatGPT Work · minor |
| Oct 13, 2026, 08:10 UTC | Degraded performance in GPT-6.1 Sol Codex and Work mode · minor |
| Oct 7, 2026, 17:21 UTC | Degraded performance in Codex and Work mode: Dot turn failures, delayed responses, and reduced proactivity · minor |
| Oct 7, 2026, 16:31 UTC | Dot in Codex may fail to create new threads · minor |
| Oct 5, 2026, 17:58 UTC | Elevated error rates in Codex Cloud · none |
OpenAI Codex FAQ
How much does OpenAI Codex cost?
Included in ChatGPT plans: Free $0, Go $8/month, Plus $20/month, Pro from $100/month ($100, $200 or $500 tiers), Business $20/user/month billed annually ($25 monthly), Enterprise and Edu by contact. API-key use is billed at API rates (no cloud features).
Is OpenAI Codex open source?
Yes. OpenAI Codex is open source, released under the Apache-2.0 license.
How popular is OpenAI Codex?
OpenAI Codex was downloaded 26.6M times from npm in the week ending Nov 17, 2026, up 2.4% on the week before and up 7.4% over the last 30 days compared with the 30 before. Its VS Code extension has 16.9M installs and average 2.7 out of 5 across 975 marketplace ratings. The GitHub repository has 136K stars, 6,347 of them added in the last 30 days.
What is the latest version of OpenAI Codex?
The latest stable release we track is 0.168.0, published on Nov 13, 2026.
What are the alternatives to OpenAI Codex?
The closest alternatives in the same category by Gid Score are Claude Code, OpenCode, Pi, Gemini CLI.
When did OpenAI last reset Codex usage limits?
On Nov 12, 2026 (UTC): Activated a banked reset that was expected to land in everyone's accounts soon, after shipping some updates to dots. Source: https://x.com/thsottiaux/status/2120999993079451919 63 resets since Sep 17, 2025, on average every 6.8 days.
For the makers of OpenAI Codex badges, corrections
Add a live badge to your README or site:
[](https://agentgid.com/agents/openai-codex/)
[](https://agentgid.com/agents/openai-codex/)
[](https://agentgid.com/agents/openai-codex/)
[](https://agentgid.com/agents/openai-codex/)
Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.
Where these numbers come from
npm: @openai/codex. GitHub: openai/codex. VS Code Marketplace: openai.chatgpt. Homebrew: codex. Wikipedia: OpenAI Codex. See data sources for how each one is collected.
Product facts and pricing were checked against: learn.chatgpt.com, registry.npmjs.org, formulae.brew.sh, marketplace.visualstudio.com, open-vsx.org, github.com, en.wikipedia.org, community.openai.com.