Devtoolwatch Boardcodex-chatgpt-web review
codex-chatgpt-web is a launcher that lets Codex use the ChatGPT Web models on your account, including Pro, through browser automation.
TL;DR
codex-chatgpt-web adds ChatGPT Web models to Codex's native model picker and sends messages through your signed-in ChatGPT session. It is for Codex users with a ChatGPT account who want those models without using Codex quota. Full harness mode connects ChatGPT to the task's files, terminal and tools through MCP.
- 15.3kGitHub stars
- +447new stars, week of Oct 25
- #7in Inference and training this week
codex-chatgpt-web at a glance
- Category
- Inference and training
- Maintained by
- vechen (individual, @miuuyy on GitHub)
- License
- MIT
- Language
- TypeScript, JavaScript, CSS
- Repository
- github.com/miuuyy/codex-chatgpt-web
- Latest release
- v6.1.17 on GitHub, Oct 29, 2026
- First published
- Jul 26, 2026
What codex-chatgpt-web does
- codex-chatgpt-web lists ChatGPT Web models in Codex's model and Effort selectors as entries ending in (Web).
- codex-chatgpt-web keeps Codex's interface, tasks, images and streaming while messages run through your ChatGPT Web session.
- codex-chatgpt-web Full harness mode connects ChatGPT tool calls to the current Codex task's files, terminal, tools and approvals through MCP.
- codex-chatgpt-web offers a Zero Risk mode where you paste and send prompts manually, and it never reads or operates the ChatGPT page.
- codex-chatgpt-web supports Bigger Context and native Codex compaction, with context limits that depend on account type and selected effort.
- codex-chatgpt-web uses Temporary Chat by default, which keeps new task conversations out of ChatGPT history.
- codex-chatgpt-web provides subagent protocols, Compatibility V1 and Native, for cross-backend delegation.
How it works
codex-chatgpt-web is a desktop launcher with an embedded browser and runtime. You sign in to ChatGPT in that browser, and the launcher installs models into Codex. In Full harness mode, the official OpenAI tunnel-client relays ChatGPT tool calls back to the Codex task over an outbound tunnel. It needs no public IP or inbound port. A ChatGPT Developer Mode connector named Codex Native2 makes the link.
How to install and start using codex-chatgpt-web
macOS / Linux
curl -fsSL https://github.com/miuuyy/codex-chatgpt-web/releases/latest/download/install-launcher.sh | shWindows PowerShell
irm https://github.com/miuuyy/codex-chatgpt-web/releases/latest/download/install-launcher.ps1 | iexRequirements
- A ChatGPT account. Available models depend on the account and workspace.
- Codex installed.
- macOS 13+ (arm64/x64), Windows x64, or Linux x64. The app includes its browser and runtime.
- Bun 1.4.0 to run from source.
First steps
- Install the codex-chatgpt-web launcher using the download or installer for your system.
- Sign in to ChatGPT in the embedded browser and run the browser smoke test.
- Install models, restart Codex once, then choose a model ending in (Web).
- For coding with tools, open MCP in the launcher and complete the Full harness setup.
Who codex-chatgpt-web is for
- Codex users who want ChatGPT Web models, including Pro, without using Codex quota.
- Developers who want ChatGPT to operate on local files and terminals through MCP from Codex tasks.
- Users on a trusted personal workstation using their own ChatGPT account.
Limitations to know
- codex-chatgpt-web is unofficial browser automation, not an OpenAI API. ChatGPT UI changes can break selectors.
- Messages are subject to your ChatGPT account's limits, and task prompts are processed by OpenAI.
- Builds are not yet platform-signed, so Gatekeeper or SmartScreen may show warnings.
- Browser state is a sensitive login artifact. The loopback listener is reachable by processes running as the same local user.
Is codex-chatgpt-web growing? Usage data
In the week of Oct 25–31, 2026, codex-chatgpt-web ranked #46 on the Devtoolwatch Board and #7 in Inference and training, with a momentum score of 77 out of 100.
Show the weekly numbers
| Week of | New stars |
|---|---|
| Aug 9, 2026 | 136 |
| Aug 16, 2026 | 425 |
| Aug 23, 2026 | 1.1k |
| Aug 30, 2026 | 2.4k |
| Sep 6, 2026 | 1.7k |
| Sep 13, 2026 | 3k |
| Sep 20, 2026 | 2.1k |
| Sep 27, 2026 | 1.6k |
| Oct 4, 2026 | 610 |
| Oct 11, 2026 | 605 |
| Oct 18, 2026 | 491 |
| Oct 25, 2026 | 447 |
The Board ranks tools by four signals each week: new stars, new stars for the repo's size, added downloads and download growth beyond the market. How the ranking works.
codex-chatgpt-web alternatives
Other LLM inference and training tools on the Board in the week of Oct 25–31, 2026, by momentum:
| Rank | Tool | Stars | Downloads/wk |
|---|---|---|---|
| #1 | dsh-our-free-modeldsh-our-free-model is a dsh plugin that provides DeepSeek V4.1 Flash, Kimi K3 and other models without login, sign-up or API key. | 54.3k | – |
| #2 | magpiemagpie is a local gateway and app that lets coding agents like Claude Code, Codex and Gemini CLI use any provider's model. | 26.9k | – |
| #3 | StrataStrata is a local inference engine and app that runs large open models like Qwen3.8-Flash-Next on consumer hardware with… | 37.1k | – |
| #4 | edge0edge0 is an open-source streaming inference framework that runs large sparse mixture-of-experts models on consumer hardware by streaming… | 8.9k | – |
| #5 | OmniRouteOmniRoute is an open-source AI gateway that routes requests across many model providers through one OpenAI-compatible endpoint, with… | 83.2k | 106k |
| #6 | CLIProxyAPICLIProxyAPI is a Go proxy server that exposes CLI coding accounts like Codex, Claude Code and Gemini CLI through OpenAI, Gemini and… | 57.1k | – |
Recent codex-chatgpt-web releases
codex-chatgpt-web FAQ
Is codex-chatgpt-web free and open source?
Yes. codex-chatgpt-web's source code is on GitHub (miuuyy/codex-chatgpt-web) under the MIT license, so you can use it for free.
How do I install codex-chatgpt-web?
Run curl -fsSL https://github.com/miuuyy/codex-chatgpt-web/releases/latest/download/install-launcher.sh | sh. The install section above has the other options from the project's README.
Is codex-chatgpt-web still maintained?
Yes. The miuuyy/codex-chatgpt-web repository was last updated on Oct 30, 2026, and the latest release, v6.1.17 on GitHub, came out on Oct 29, 2026.
How popular is codex-chatgpt-web?
codex-chatgpt-web has 15.3k GitHub stars as of Nov 1, 2026. It gained 447 new stars in the week of Oct 25–31, 2026, against a 12-week average of 1.2k a week. That made it #7 among LLM inference and training tools on the Devtoolwatch Board.
What are the best alternatives to codex-chatgpt-web?
Other LLM inference and training tools on the Devtoolwatch Board in the week of Oct 25–31, 2026, by momentum: dsh-our-free-model (#1), magpie (#2), Strata (#3), edge0 (#4), OmniRoute (#5). The alternatives table above has their numbers.
Does codex-chatgpt-web use my Codex quota?
No. codex-chatgpt-web sends messages through your signed-in ChatGPT Web session, so they count against that ChatGPT account's limits instead of Codex quota.
Which models does codex-chatgpt-web offer?
codex-chatgpt-web offers Luna / Think on Free and Go accounts. Paid accounts get GPT-6 Sol and GPT-5.6 Sol, with Instant–High, plus Extra High and Pro when available. Availability depends on your account and workspace.
Does the Full harness tunnel open ports on my machine?
No. The codex-chatgpt-web tunnel is outbound. It does not expose a public IP, open an inbound port, or require router forwarding. It uses your system proxy.
What should I do if Codex shows only Luna (Web) on a paid account?
Install codex-chatgpt-web 6.1.16 or later. Then open Setup → Install into Codex → Reinstall and fully restart Codex. Reinstalling models on 6.1.15 cannot repair this bug.
Sources
codex-chatgpt-web README badge
Maintainers can show the tool's weekly Board rank in their README:
[](https://index.devtoolwatch.com/tools/miuuyy-codex-chatgpt-web/)How this page is made: Devtoolwatch writes each review from the project's README, releases, package metadata and website, with an AI model, then checks every claim against the source it cites and drops what the sources don't support. The numbers come from Devtoolwatch's daily GitHub snapshots, GitHub's stargazer history, and the npm and PyPI download APIs. Spotted an error? Email [email protected].