5.4, in my own testing, was almost always ahead of Opus 4.6 for reviews and planning. I'm on plus plan on openai, so I couldn't test it so deeply. Anyone who had more experience on both could perhaps chime in? Pros/cons compared to Opus? I'm invested in Claude ecosystem but the recent quality and session limits decrease have me on the edge.
Same for me. I'm on $20 plan for both and I use them both interchangeably. Similar "intelligence" imo. Just different way of doing things, that's all. But Claude is getting worse in terms of token usage so I've cancelled my plan last month.
I use opencode, so can toggle between Claude and Codex fairly easily, and do so whenever one of them is having problems (until yesterday, that is, when Claude blocked opencode for good, and I cancelled my account). This means I'm using the same prompts and instructions for both.
Personally, it seems like I have to redirect Opus/Sonnet much less often. GPT felt pretty "dense", it was more likely to ignore earlier instructions in the session, I had to remind it more often, and when I reviewed the code it produced I had to make more corrections that seemed obvious.
Entirely subjective, but I also find I prefer Claude's "personality" to ChatGPT, but I couldn't point to any specific differences.
Yeah it's probably a bit better overall. 5.4 is a month newer than Opus 4.6
My guess is that 5.5 will come out soon and be significantly better so you'd want to be using Codex then, but then when Opus 5 comes out probably back to claude code
Also 5.4 has fast mode, and higher usage limits since it's cheaper
Just curious as I've often heard that Claude was superior for planning/architecture work while ChatGPT was superior for actual implementation and finding bugs.
Claude makes more detailed plans that seem better if you just skim them, but when analyzed, has a lot of errors, usually.
It compensates for most during implementation if you make it use TDD by using superpower et al, or just telling it to do so.
GPT 5.4 makes more simple plans (compared to superpowers - a plugin from the official claude plugin marketplace - not the plan mode), but can better fill the details while implementing.
Plan mode in Claude Code got much better in the last months, but the lacking details cannot be compensated by the model during the implementation.
So my workflow has been:
Make claude plan with superpowers:brainstorm, review the spec, make updates, give the spec to gpt, usually to witness grave errors found by gpt, spec gets updates, another manual review, (many iterations later), final spec is written, write the plan, gpt finds mind boggling errors, (many iterations later), claude agent swarm implements, gpt finds even more errors, I find errors, fix fix fix, manual code review and red tests from me, tests get fixed (many iterations later) finally something usable with stylistic issues at most (human opinion)!
This happens with the most complex features that'd be a nightmare to implement even for the most experienced programmers of course. For basic things, most SOTA modals can one-shot anyway.
Interesting. Have you ever had Claude re-review its plan after having it draft the original plan? Or do you give it to GPT right away to review?
Just curious as I'm trying to branch out from using Claude for everything, and I've been following a somewhat similar workflow to yours, except just having Claude review and re-review its plan (sometimes using different roles, e.g. system architect vs SWE vs QA eng) and it will similarly identify issues that it missed originally.
But now I'm curious to try this while weaving in more GPT.
Draw a line, say this is for bicycles, pedestrians and cars have no business here, and bikes have no business being on any other lane as long as these exist.
When bikes have to go through areas where people walk freely, they need to limit their speed to a walking pace.
People should not wear headphones (noise-cancelling or not) when going through traffic as pedestrians. Take them off when crossing!
People should not hear loud music when driving - max is normal speaking voice level. Bike drivers should never hear any music, let alone wearing headphones. Behind-ear speakers on low could be a compromise.
> Draw a line, say this is for bicycles, pedestrians and cars have no business here, and bikes have no business being on any other lane as long as these exist.
This is the reality in many cities, if it weren't for the hopefully not surprising fact that people don't always obey traffic laws perfectly.
No, you didn't. And restricting cyclists and pedestrians will not result in even small dent in the numbers of maimed or killed people in traffic. It's one mode of transport that's responsible for the vast amount of it, and that's the motorized one propelling several tonnes.
> and bikes have no business being on any other lane as long as these exist
And cars have no business being on other roads as long as highways exist ;)
I meant biking accidents this product is obviously trying to solve.
> And cars have no business being on other roads as long as highways exist ;)
Biking lanes are not comparable to highways. Where I'm living, if you bike on car lanes when biking lanes exist, or if you bike on sidewalks at all, you get a hefty fine depending on the situation and if you possess one, you get points on your driving license.
Exceptions are turning, leaving the road, the lane being blocked by a clueless driver etc. obviously.
Cars are also not allowed on biking lanes, neither are pedestrians. Same exceptions apply.
How do we enforce seatbelts? (1) Assume the public aren't stupid. (2) Assume the public aren't murderers. (3) Explain the risk-benefit analysis through informative videos like https://en.wikipedia.org/wiki/Julie_(1998_film).
People can shout "domestic terror" all they like, but if it's not true, it's not true.
Wearing a seatbelt cost next to nothing in inconvenience. Not being able to listen to music or have phone calls with noise cancellation while walking does not really compare.
Of course this requires compensating for the loss in awareness through hearing by looking more diligently before crossing a bike lane, but unfortunately, some people never learn this, or only through a few close calls.
"Annoyingly" ringing a bell and converting a potential accident into a close call seems pretty close to optimal to me.
"Next to nothing in inconvenience" is the perception now. It certainly wasn't the perception when seatbelts were introduced. The ability to listen to personal music while walking is less than 50 years old: before that, you had the radio or nothing. Even that would not be an intolerable inconvenience for most. But I was more thinking:
> People should not hear loud music when driving - max is normal speaking voice level.
which feels like a more than acceptable constraint to me.
> People should not hear loud music when driving - max is normal speaking voice level.
Oh, completely agreed on that one. In a car, you are also by far better protected than any cyclists you might encounter, so you shouldn't make it harder to hear their signaling. (I still wouldn't rely on any car having heard my bell if I don't get any other confirmation that the driver has noticed me, e.g. sufficiently slowing down as they are approaching the intersection where I have right of way.)
But GGP also said
> People should not wear headphones (noise-cancelling or not) when going through traffic as pedestrians. Take them off when crossing!
and that's what I think goes too far. Why should I remove my headphones if I look both ways before crossing a bike lane or road?
The ideal rule would of course be that only those pedestrians remove their headphones that are otherwise inattentive... Although I have my doubts that they'd remember.
You are answering different question. What you are saying is called awareness campaign or something. Enforcement of seatbelts is done by police with fines/tickets and is possible cause it's visible from outside.
Other things like loudness levels inside cars cannot be monitored without going in full totalitarian mode.
Why would enforcement be necessary, given assumptions 1 and 2 (not stupid, not murderers), and awareness? Around these parts, seatbelt enforcement isn't necessary because everyone voluntarily wears their seatbelt – except for children, occasionally, but the adults are generally capable of enforcing that. (Even teenagers / young adults being irresponsible in cars generally wear seatbelts while doing so.)
Cause humans works that way. We don't calculate risks based on statistics in everyday life. That means you can't just make people aware, you need to build cultural norm around that. But cultural norms without enforcement are easy to erode. When norm is systematically broken it ceases to be norm.
>seatbelt enforcement isn't necessary because everyone voluntarily wears their seatbelt
You obviously don't travel much. There are whole countries where seatbelts are rarely used, and drivers buy special plugs for belt receptacles to silence beeping.
Unfortunately, the UK seems almost incapable of building usable cycle infrastructure (possibly excepting London). Your idea is just a recipe for magic protective paint and even more abuse of cyclists who don't want to be forced to use ridiculously badly designed infrastructure. e.g. Here in Bristol, we have an infamous shared cycle/pedestrian pavement along Coronation Rd that has a few trees completely blocking the cycle side which just means conflict between pedestrians and cyclists who have to fight over the scraps left over from motorists taking most of the space (https://www.google.com/maps/@51.4462522,-2.6064792,3a,75y,80...).
If "shouldn't" worked we'd have no industrial accidents without any safety measures, no unwanted pregnancies and in general would more or less achieve heaven on Earth.
> Becker argues that a basic duality in human life exists between the physical world of objects and biology, and a symbolic world of human meaning. Thus, since humanity has a dualistic nature consisting of a physical self and a symbolic self, we are able to transcend the dilemma of mortality by focusing our attention mainly on our symbolic selves, i.e. our culturally based self esteem, which Becker calls "heroism": a "defiant creation of meaning" expressing "the myth of the significance of human life" as compared to other animals. This counters the personal insignificance and finitude that death represents in the human mind.
> Such symbolic self-focus takes the form of an individual's "causa sui project", (sometimes called an "immortality project", or a "heroism project"). A person's "causa sui project" acts as their immortality vessel, whereby they subscribe to a particular set of culturally-created meanings and through them gain personal significance beyond that afforded to other mortal animals. This enables the individual to imagine at least some vestige of those meanings continuing beyond their own life-span; thus avoiding the complete "self-negation" we perceive when other biological creatures die in nature.
You can find big similarities such as the promised land as the immortality vessel, heroism as a response to historical trauma and the ongoing attacks on their sovereignty, and the immortality project would be the nation-state. Becker goes on to categorize all of this similar to a mental illness. You can read the wikipedia page here, I find it very helpful: https://en.wikipedia.org/wiki/The_Denial_of_Death
TL;DR: If you look through Becker's lens, you start to realize how stupid such wars and expansionist ideals seem. People should focus on what exists now and stop chasing projects that'd span beyond their lifetimes while making life today worse.
The book doesn't support anything in "their" favor. The idea boils down to nobody being "them".
Related part from Wikipedia:
> Becker argues that the conflict between contradictory immortality projects (particularly in religion) is a major source of the violence and misery in the world such as wars, genocide, racism, nationalism and so forth since immortality projects that contradict one another threaten one's core beliefs and sense of security
What people also underestimate is the new power of the index.php that comes from the LLMs.
Tell claude to create a php backend to your portfolio html template, drag the generated file to the cheapest server, and you already have a custom CMS.
But in search results, you only find the sites that game the system to maximize their profits, while millions of other well-meaning sites get little to no traffic, and eventually people lose interest in maintaining an online presence. They move toward big silos like Instagram, platforms that just use their content to attract more ads.
Ads do break the internet, or let's say, fundamentally change the model of how it works to the detriment of most people
reply