News for hackers. Stuff that matters.
Re:System Card: Claude Mythos Preview [pdf] by rafaelmn on Tuesday April 07, 2026 @02:56PM (#47679757)
GPT is shit at writing code. It's not dumb - extra high thinking is really good at catching stuff - but it's like letting a smart junior into your codebase - ignore all the conventions, surrounding context, just slop all over the place to get it working. Claude is just a level above in terms of editing code.
Re:System Card: Claude Mythos Preview [pdf] by Jcampuzano2 on Tuesday April 07, 2026 @03:03PM (#47679851)
Not my experience. GPT 5.4 walks all over Claude from what I've worked with and its Claude that is the one willing to just go do unnecessary stuff that was never asked for or implement the more hacky solutions to things without a care for maintainability/readability.

But I do not use extra high thinking unless its for code review. I sit at GPT 5.4 high 95% of the time.

Re:System Card: Claude Mythos Preview [pdf] by sho_hn on Tuesday April 07, 2026 @03:22PM (#47680123)
Very different experience for me. Codex 5.3+ on xhigh are the only models I've tried so far that write reasonably decent C++ (domains: desktop GUI, robotics, game engine dev, embedded stuff, general systems engineering-type codebases), and idiomatic code in languages not well-represented in training data, e.g. QML. One thing I like is explicitly that it knows better when to stop, instead of brute-forcing a solution by spamming bespoke helpers everywhere no rational dev would write that way.

Not always, no, and it takes investment in good prompting/guardrails/plans/explicit test recipes for sure. I'm still on average better at programming in context than Codex 5.4, even if slower. But in terms of "task complexity I can entrust to a model and not be completely disappointed and annoyed", it scores the best so far. Saves a lot on review/iteration overhead.

It's annoying, too, because I don't much like OpenAI as a company.

(Background: 25 years of C++ etc.)

Re:System Card: Claude Mythos Preview [pdf] by boring-human on Tuesday April 07, 2026 @05:09PM (#47681410)
Same background as you, and same exact experience as you. Opus and Gemini have not come close to Codex for C++ work. I also run exclusively on xhigh. Its handling of complexity is unmatched.

At least until next week when Mythos and GPT 6 throw it all up in the air again.

Re:System Card: Claude Mythos Preview [pdf] by zarzavat on Tuesday April 07, 2026 @03:05PM (#47679878)
Yes, it's becoming clear that OpenAI kinda sucks at alignment. GPT-5 can pass all the benchmarks but it just doesn't "feel good" like Claude or Gemini.
Re:System Card: Claude Mythos Preview [pdf] by lilytweed on Tuesday April 07, 2026 @03:17PM (#47680042)
Whenever I come back to ChatGPT after using Claude or Gemini for an extended period, I’m really struck by the “AI-ness.” All the verbal tics and, truly, sloppishness, have been trained away by the other, more human-feeling models at this point.
Re:System Card: Claude Mythos Preview [pdf] by kranke155 on Tuesday April 07, 2026 @05:59PM (#47681862)
GPT was clearly changed after its sycophantic models lead to the lawsuits.
Re:System Card: Claude Mythos Preview [pdf] by josephg on Tuesday April 07, 2026 @07:10PM (#47682497)
It still has a very ... plastic feeling. The way it writes feels cheap somehow. I don't know why, but Claude seems much more natural to me. I enjoy reading its writing a lot more.

That said, I'll often throw a prompt into both claude and chatgpt and read both answers. GPT is frequently smarter.

Re:System Card: Claude Mythos Preview [pdf] by kranke155 on Wednesday April 08, 2026 @06:14AM (#47688045)
GPT is more accurate. But Claude has this way of association between things that seems smarter and more human to me.
Re:System Card: Claude Mythos Preview [pdf] by chaos_emergent on Tuesday April 07, 2026 @03:32PM (#47680246)
An alternative but similar formulation of that statement is that Anthropic has spent more training effort in getting the model to “feel good” rather than being correct on verifiable tasks. Which more or less tracks with my experience of using the model.
Re:System Card: Claude Mythos Preview [pdf] by zarzavat on Wednesday April 08, 2026 @07:50AM (#47688900)
Alignment is a subspace of capability. Feeling good is nice, but it's also a manifestation of the level that the model can predict what I do and don't want it to do. The more accurately it can predict my intentions without me having to spell them out explicitly in the prompt, the more helpful it is.

GPT-5 is good at benchmarks, but benchmarks are more forgiving of a misaligned model. Many real world tasks often don't require strong reasoning abilities or high intelligence, so much as the ability to understand what the task is with a minimal prompt.

Not every shop assistant needs a physics degree, and not every physics professor is necessarily qualified to be a shop assistant. A person, or LLM, can be very smart while at the same time very bad at understanding people.

For example, if GPT-5 takes my code and rearranges something for no reason, that's not going to affect its benchmarks because the code will still produce the same answers. But now I have to spend more time reviewing its output to make sure it hasn't done that. The more time I have to spend post-processing its output, the lower its capabilities are since the measurement of capability on real world tasks is often the amount of time saved.

Re:System Card: Claude Mythos Preview [pdf] by leobuskin on Tuesday April 07, 2026 @03:02PM (#47679842)
And as a bonus: GPT is slow. I’m doing a lot of RE (IDA Pro + MCP), even when 5.4 gives a little bit better guesses (rarely, but happens) - it takes x2-x4 longer. So, it’s just easier to reiterate with Opus
Re:System Card: Claude Mythos Preview [pdf] by blazespin on Tuesday April 07, 2026 @04:48PM (#47681177)
Yeah, need some good RE benchmarks for the LLMs. :)

RE is very interesting problem. A lot more that SWE can be RE'd. I've found the LLMs are reluctant to assist, though you can workaround.

Re:System Card: Claude Mythos Preview [pdf] by porker on Tuesday April 07, 2026 @05:02PM (#47681340)
What is RE in this context?
Re:System Card: Claude Mythos Preview [pdf] by astrange on Tuesday April 07, 2026 @05:16PM (#47681473)
Reverse engineering
Re:System Card: Claude Mythos Preview [pdf] by 19h on Tuesday April 07, 2026 @07:53PM (#47682852)
Mind sharing the use cases you're using IDA via MCP for?
Re:System Card: Claude Mythos Preview [pdf] by aizk on Tuesday April 07, 2026 @09:57PM (#47683908)
I've been messing with using Claude, Codex, and Kimi even for reverse engineering at https://decomp.dev/ it's a ton of fun. Great because matching bytes is a scoring function that's easy for the models to understand and make progress on.
Re:System Card: Claude Mythos Preview [pdf] by adamgoodapp on Wednesday April 08, 2026 @03:34AM (#47686653)
I want to get into RE with AI. Which model you liking the most?
Re:System Card: Claude Mythos Preview [pdf] by rustystump on Wednesday April 08, 2026 @11:20AM (#47691462)
This. People drastically underestimate how much more useful a lightning fast slightly dumb model is compared to a super smart but mega slow model is. Sure, u may need to bust out the beef now and then. However, the overwhelming majority of work the fast stupid model is a better fit.
Re:System Card: Claude Mythos Preview [pdf] by whalesalad on Tuesday April 07, 2026 @02:59PM (#47679805)
This has been my experience. With very very rigid constraints it does ok, but without them it will optimize expediency and getting it done at the expense of integrating with the broader system.
Re:System Card: Claude Mythos Preview [pdf] by ctoth on Tuesday April 07, 2026 @03:35PM (#47680290)
My favorite example of this from last night:

Me: Let's figure out how to clone our company Wordpress theme in Hugo. Here're some tools you can use, here's a way to compare screenshots, iterate until 0% difference.

Codex: Okay Boss! I did the thing! I couldn't get the CSS to match so I just took PNGs of the original site and put them in place! Matches 100%!

Re:System Card: Claude Mythos Preview [pdf] by camdenreslink on Tuesday April 07, 2026 @06:05PM (#47681956)
ChatGPT 5.4 with extra high reasoning has worked really well for me, and I don't notice a huge difference with Opus 4.6 with high reasoning (those are the 2 models/thinking modes I've used the most in the last month or so).