# GPT-5.2-Codex

54 Hacker News commenters recommended GPT-5.2-Codex and 33 criticized it, November 2025 to November 2026. It was recommended in 4 of 13 months.

**Verdict:** People recommend it for strong coding, bug finding and code review. Common caveat: it is not clearly the best across the board and makes design errors.

## Why people recommend it

- Strong at finding bugs and issues (5 comments): "Codex is not only finding much higher quality issues, it's also writing code that usually doesn't leave quality issues behind." ([source](https://news.ycombinator.com/item?id=46321028)) "GPT-5.2-Codex still get 0% false positive, while getting at the level of other best models for localization of backdoors" ([source](https://news.ycombinator.com/item?id=47158365)) "It took a couple minutes, but it managed to track the issue down using a bunch of tricks I've never seen before." ([source](https://news.ycombinator.com/item?id=46320821))
- Competitive with Opus 4.5 (5 comments): "GPT 5.2 and the codex variants are, to me, every bit as good as Opus but without the groveling and emojis" ([source](https://news.ycombinator.com/item?id=46519702)) "GPT 5.2 (-codex) in Codex, which is actually far more capable at such work than Opus 4.5 in Claude Code" ([source](https://news.ycombinator.com/item?id=46791210)) "you'll have better odds with Codex GPT-5.2-Codex-XHigh than with Claude Code Opus 4.5" ([source](https://news.ycombinator.com/item?id=46791686))
- Clear improvement over earlier models (4 comments): "the big unlock was GPT-5.2 and GPT-5.2-Codex, where we found ourselves needing to make many fewer manual edits" ([source](https://news.ycombinator.com/item?id=46859713)) "5.2 Codex are both so clearly more competent than previous models" ([source](https://news.ycombinator.com/item?id=46438585)) "Codex with 5.0 and 5.1 wasn't great, but 5.2 is big improvement." ([source](https://news.ycombinator.com/item?id=46318017))
- High usage limits on cheap plans (3 comments): "it has actually gotten much better, faster and the most important one, more time before you reach your weekly limit" ([source](https://news.ycombinator.com/item?id=46495989)) "5.2-codex is pretty solid and you get dramatically higher usage rates with cheap plans" ([source](https://news.ycombinator.com/item?id=46880135)) "I'm never running into limits, and I'm running these models near all day every day" ([source](https://news.ycombinator.com/item?id=46316789))

## What people warn about

- Makes mistakes and breaks code (6 comments): "I find that 5.2 overfits to the more common implementation and will actively break working code and puzzles." ([source](https://news.ycombinator.com/item?id=46316590)) "It seems to have become significantly worse in the past couple of days. It keeps making dumb mistakes" ([source](https://news.ycombinator.com/item?id=46893669)) "they fail in strange ways and if you're well versed in your domain, these are easy to spot." ([source](https://news.ycombinator.com/item?id=46983153))
- Goes in circles and runs slow (4 comments): "added hundreds of lines of code several times, and for 15-20 iterations I did not change one thing" ([source](https://news.ycombinator.com/item?id=46869131)) "Generally I have noticed Gpt 5.2 codex is slower compared to Sonnet 4.5 in Claude Code." ([source](https://news.ycombinator.com/item?id=46738802)) "GPT 5.2 codex plans well but fucks off a lot, goes in circles (more than opus 4.5)" ([source](https://news.ycombinator.com/item?id=46905045))
- Not clearly better than rivals (3 comments): "GPT-5.2-Codex is strong but not clearly the new state of the art across the board" ([source](https://news.ycombinator.com/item?id=46324717)) "With OpenCode I also switched to GPT-5.2-Codex and I have to say it's fairly garbage IME" ([source](https://news.ycombinator.com/item?id=46858979)) "OpenAI just straight up doesn't compare their model to others on these release pages" ([source](https://news.ycombinator.com/item?id=46316993))

## Alternatives hackers recommend

- [Cursor Composer](https://hackerapproved.com/p/composer-2-5.md): 214 recommend
- [Qwen3-Coder](https://hackerapproved.com/p/qwen3-coder-next.md): 165 recommend
- [GPT-5.3-Codex](https://hackerapproved.com/p/gpt-5-3-codex.md): 135 recommend
- [Meta Muse Glimmer 30B](https://hackerapproved.com/p/meta-muse-glimmer.md): 70 recommend
- [Laguna S 2.1](https://hackerapproved.com/p/laguna-s-2-1.md): 43 recommend

## Where to get it

- [Visit the official site](https://openai.com/)

## Cite as

Hacker Approved, "GPT-5.2-Codex", https://hackerapproved.com/p/gpt-5-2-codex/ (counts from 1,795,640 Hacker News comments, November 2025 to November 2026).