Hacker News Reader

Comment by wxw | Hacker News Reader

I’ve been using Codex w GPT 5.5 more than Claude Code recently. I think Anthropic won the marketing game because Codex is quite good, even better IME.

View on HN

I started using GPT for coding for the first time this week, and I’m sort of in awe at how well it’s following all of my hooks, skills, and prompts. GPT is so unlazy and deterministic, it’s honestly been so refreshing this week. I cancelled my Claude 20X and replaced it with GPT 20X, no regrets so far.

That said, I feel icky, like I just made a Facebook account in 2026 :(

I'm not sure about that. Claude has some bugs, but Codex is not as polished and doesn't have as many features. For example, you need to add MCP servers manually. There's no Plugin/Skill/Connector marketplace that is accessible from within the app, like there is with Claude Desktop. The Cowork-equivalent is nowhere as powerful. And so on.

I still use Codex, but mostly when I need to check Opus 4.8's work. Pretty sure I will stop doing that soon, because during the short time Fable was available, Codex was not able to find any important issues with the code Fable wrote.

Personally I prefer GPT 5.5 writing style over Opus 4.8. It’s much more no nonsense and information denser.

That's the first time I saw someone prefering GPT-styled output over Claude ;) It's the complete opposite for me, GPT is way too verbose (even after telling it to STFU), overwhelms the user with thousands of options and doesn't just answer a question without shitting out thousands of paragraphs. Also the overall tone is way too enthusiastic.

Dunno, I prefer GPT 5.5 too for the same reasons as the parent. Extremely subjective but had better results with it too. Maybe I just got unlucky with Claude a few times, but even the latest Opus was dumb.

Fascinating how people have such complete diametrically opposed experiences. I guess both models have it in them to behave very differently in different circumstances and we have very little idea what pushes them in this or that direction. I guess it does boil down to luck!

Personally, Claude Opus (and in the few interactions I had with it, Fable) has been the far the superior experience. GPT-5.5 seems dumber and more certain about presenting me bullshit. Opus has better humor, and is less pretentious in its presentation. But this may all boil down to how the models react to my prompting.

What is without a doubt is that I wish they both were more intelligent – or maybe it is their wisdom I find lacking!

It might be harness or prompting style. Personally I use opencode and my prompting style is very plain and terse . Where tasks are very small. Opus and Sonet too often are too verbose and go tangent. Where GPT5.5 is much stricter.

I strongly prefer codex. Claude is annoying. Codex provides descriptions where I want them and more touchpoints to audit the quality of work. Claude code on experimental seems to not even show diffs when asked anymore, and it's much less clear what is being shipped.

There are plugins in the app.

Haven’t tried Cowork, interesting. Isn’t it just the same agent minus the git worktree based UI?

Frankly, neither Claude nor Codex are as good as hype entails.

But how many plugins are people actually using? I can think of one MCP server I find valuable (context7) and one plugin that i've installed, but continuously think about uninstalling (obra/superpowers).

Both were trivial to set up with codex.

> For example, you need to add MCP servers manually. There's no Plugin/Skill/Connector marketplace that is accessible from within the app

This is all wrong.

i think codex is much better in that aspect. In claude there is skills, connector, capabilities and 4 places for browser... It is too much.

It's a good thing. I hate MCPs from the bottom of my heart because they always stay there and bloat the context window. Also, usually developers who develop them don't know what they're doing, so the MCP responses also bloat your context even further.

Claude has always been better at making pretty frontends, which is crucial for people that vibecode entire apps in a couple of prompts. And that people drive a lot of the hype.

Codex ever since ~5.2 has been better at long tasks in large codebases.

One very annoying thing about Codex: You can't do anything until all MCP servers are loaded and connected (or failed).

Can you move it to background connection?

Unless the MCPs is connected through STDIO, I see no reason for them to block Codex from accepting user prompt. I know other harnesses doesn’t wait for all MCPs to connect. They can establish their connection during a session, they’re dynamic after all.

Reply on news.ycombinator.com