# OpenAI Codex 2474 Hacker News commenters recommended OpenAI Codex and 1289 criticized it, November 2025 to November 2026. It was recommended in 13 of 13 months; owners who said how long report a median of 0.25 years (165 owners). **Verdict:** A coding agent many find as good as or better than Claude Code, thorough on hard problems, with generous $20-plan limits, though its output needs careful review. ## Why people recommend it - Rivals or beats Claude Code (6 comments): "Codex is good now. I'm undecided which is better, but they're definitely close enough that I feel comfortable recommending Claude-exclusive people in my circle to try codex." ([source](https://news.ycombinator.com/item?id=48148780)) "codex's remote session integration is better, so outside of some "ultracode" orchestration bells/whistles where Claude Code is ahead, I think Codex is a better tool" ([source](https://news.ycombinator.com/item?id=48849156)) "codex is currently a much better model, and it also seems to be technically stronger on the type of applied math and theory heavy programming I like to do" ([source](https://news.ycombinator.com/item?id=49104582)) - Handles laborious tasks and tooling (6 comments): "With a rather short prompt, claude/codex will take your code, write a harness, profile it, build experiments, profile those, and give some pretty solid advice which one to pick." ([source](https://news.ycombinator.com/item?id=49468735)) "Claude Code/Codex are useful for doing certain laborious tasks like bookkeeping errors/reconciliation issues but they don't replace a professional accountant." ([source](https://news.ycombinator.com/item?id=48133962)) "codex did my taxes this year (well it actually implemented a normalization pipeline and a tax computing engine which then did the taxes, but close enough)" ([source](https://news.ycombinator.com/item?id=47797078)) - Thorough and persistent on hard problems (4 comments): "codex is currently a much better model, and it also seems to be technically stronger on the type of applied math and theory heavy programming I like to do" ([source](https://news.ycombinator.com/item?id=49104582)) "Codex is better at this because it keeps going at hard problems. But then it compacts so many times over that it forgets the todo instructions." ([source](https://news.ycombinator.com/item?id=46844394)) "Codex reviewed my immich setup and recommended I change things the clip model, came up with a better backup strat and found holes in my system." ([source](https://news.ycombinator.com/item?id=49541723)) - Generous usage on $20 plan (3 comments): "I use Codex CLI in my daily usage since just with my $20/month subscription to ChatGPT, I never gets close to the quota. But it trips up over itself every now and then." ([source](https://news.ycombinator.com/item?id=47175772)) "The Codex plan for the $20 ChatGPT plan goes much further than Claude's $20 plan, but it's still not enough if you plan to work full-time with it." ([source](https://news.ycombinator.com/item?id=47541261)) "I'm switching because of the higher usage limits, 2x speed mode that isn't billed as extra usage, and much more stable and polished Mac app." ([source](https://news.ycombinator.com/item?id=47796693)) ## What people warn about - Output needs supervision and verification (4 comments): "Most often with medium to high complexity problems, if the AI comes up with a working solution it has enough problems that an attentive human reviewer would have rejected it at best." ([source](https://news.ycombinator.com/item?id=48139690)) "The agent model sounds powerful, but in practice it's still a lot of supervision, retries, and quiet hope it doesn't hallucinate itself into a refactor I didn't ask for." ([source](https://news.ycombinator.com/item?id=46860423)) "Codex has had none of these problems. I still don't trust anything it produces, but it's not like everything it produces is completely and utterly useless." ([source](https://news.ycombinator.com/item?id=47795076)) - Gets stuck or trips up (3 comments): "I use Codex CLI in my daily usage since just with my $20/month subscription to ChatGPT, I never gets close to the quota. But it trips up over itself every now and then." ([source](https://news.ycombinator.com/item?id=47175772)) "This 'getting stuck once every ten-fifteen minues' is very much the experience trying to develop complex software with codex or Claude code right now." ([source](https://news.ycombinator.com/item?id=48357989)) "I figured I'd just have Codex do it. I realized after about 3 days of various back and forth in between other work things that we were somehow stuck." ([source](https://news.ycombinator.com/item?id=49248726)) - Auto-compaction loses context (2 comments): "Codex is better at this because it keeps going at hard problems. But then it compacts so many times over that it forgets the todo instructions." ([source](https://news.ycombinator.com/item?id=46844394)) "The fact there is no way to disable auto-compaction and no way to go back in the conversation history to before a compact makes codex a no-go for me on any codebase > 5kloc." ([source](https://news.ycombinator.com/item?id=48969402)) - Usage limits too tight (2 comments): "The Codex plan for the $20 ChatGPT plan goes much further than Claude's $20 plan, but it's still not enough if you plan to work full-time with it." ([source](https://news.ycombinator.com/item?id=47541261)) "What bothers me about these cuts is the assumption that because the models are supposedly getting better and cheaper, less usage is somehow equivalent value." ([source](https://news.ycombinator.com/item?id=49895691)) ## Alternatives hackers recommend - [Claude Code](https://hackerapproved.com/p/claude-code.md): 4522 recommend - [OpenCode](https://hackerapproved.com/p/opencode.md): 1060 recommend - [Gemini CLI](https://hackerapproved.com/p/gemini-cli.md): 172 recommend - [Cline](https://hackerapproved.com/p/cline.md): 63 recommend - [Mistral Vibe](https://hackerapproved.com/p/mistral-vibe.md): 57 recommend ## Where to get it - [Visit the official site](https://openai.com/codex/) ## Cite as Hacker Approved, "OpenAI Codex", https://hackerapproved.com/p/codex/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).