GPT-5-Codex • Hacker News Board

Re: GPT-5-Codex

#141

Post by drewbitt » Wed, Sep 17, 2025, 11:23 PM UTC

king_magic wrote: post #139

hayd wrote: post #135

Earlier quoted context omitted.

The question is which version of node, not npm.

It doesn't matter. The install was broken as was written. That's poor hygiene.

It does matter.

Re: GPT-5-Codex

#142

Post by sleepnow » Fri, Sep 19, 2025, 4:59 PM UTC

king_magic wrote: post #37

Doesn't seem ready for prime-time. I'll be impressed when it actually installs. npm ERR! code 1 npm ERR! path /usr/local/lib/node_modules/@openai/codex/node_modules/@vscode/ripgrep npm ERR! command failed npm ERR! command sh -c node ./lib/postinstall.js npm ERR! /usr/local/lib/node_modules/@openai/codex/node_modules/@vscode/ripgrep/lib/download.js:199 npm ERR! zipFile?.close();

this should resolve that for you:

unset GITHUB_TOKEN

npm install --global @openai/codex

Re: GPT-5-Codex

#143

Post by king_magic » Mon, Sep 22, 2025, 8:56 AM UTC

drewbitt wrote: post #141

king_magic wrote: post #139

Earlier quoted context omitted.

It doesn't matter. The install was broken as was written. That's poor hygiene.

It does matter.

Why yes, you're right, poor code hygiene and poor documentation hygiene does matter. We finally agree.

Re: GPT-5-Codex

#144

Post by itbcharles » Mon, Sep 22, 2025, 4:51 PM UTC

jumploops wrote: post #28

siva7 wrote: post #22

Earlier quoted context omitted.

So you're all saying suddenly codex cli w gpt 5 codex is better than claude code? Hard to believe

Not suddenly, it's been better since GPT-5 launched. Prompting is different, but in a good way. With Claude Code, you can use less prompting, and Claude will get token happy and expand on your request. Great for greenfield/vibing, bad for iterating on existing projects. With Codex CLI, GPT-5 seems to handle instructions much more precisely. It won't just go off on it's own and do a bunch of work, it will do what you…

So is gpt-5-codex-high better than gpt-5-high? I find it hard to believe that with half the instruction tokens a MOE model could significantly outperform, but I am certainly not knowledgable in fine-tuning.