I never want to hear from developers again that they are not susceptible to marketing. I see meet ups specifically about Claude often.

Modern tupperware party.

A colleague was convinced Claude is better so we played a game. We used the claude code and codex harness and I implemented some prs they needed with gpt5.5 and opus4.7 and asked them to identify which came from which only from the code.

Couldn’t tell.

Edit: i bet 99% of people here, if presented with a test where i gave 5 models but all of the results came from one, would not be able to discern this. Just vibes all the way down.

I think Sam Altman is an asshole and I prefer to spend my money elsewhere.

Frontier models being commoditize is inevitable. OpenAI thinks they're still competing on technology, and not user experience and market reputation otherwise they'd understand the continuous negative PR generated by Altman's chaos is going to cost them everything.

Everything is great. However, I'm very concerned about a trend I'm noticing — Anthropic's API pricing keeps going up significantly. For third-party resale services, the price is simply too high.

As a result, their own internal products — like their low-code tools with subscriptions — naturally win on price. It's not exactly unfair, but it's frustrating that there are no real alternatives. It's similar to the current phone market: you have iOS/iPhone on one side and Android on the other, each at different price points.

As for alternative models to Opus:

- *Gemini Flash 3.5* — they say it's great, but in practice it's not that impressive. Results feel repetitive, and while the UI looks nice, it falls short. - *Gemini Pro* — not even worth considering. - *Groq* — hasn't come far enough, not very usable. - *Chinese models* — okay, very cheap and quite good. However, the latest *DeepSeek Pro V4* is decent in principle, but it's terribly slow and doesn't support images — no multimodality at all, which makes it essentially useless for our purposes.

So that's where things stand. We're waiting for more democratized models that are actually viable to use. A dollar per request with Anthropic is a nightmare.

This is an absolute joke.

Anthropic capitalized upon a brief window of being more code-focused, which turned into enterprize contracts.

Then on renewal rug-pulled those same enterprises - going from your seat includes all the usage a user would reasonably need, to being you pay for the seat + all tokens at API pricing. (Which they raised by how many times in a year? I don't know the actual number.)

Revenue spikes like crazy through basically hostage taking made possible by Sonnet 3.5 era sentiment + enterprise purchasing lag.

Parlay the revenue spike into the valuation.

Crazy. Those same enterprises will get sticker shock and leave. Absurd short-term thinking.

OpenAI is the better company (transparency, open sourcing things, how they handle things in general e.g. OpenClaw, how they compete, etc.) and they have the vastly better brand, the better consumer presence, and (for me and many others) they have the better coding app + models.

Anthropic doing deeply customer hostile stuff - again and again - to produce a short term revenue spike does NOT make for a long-term sustainable business.

For such a young business to have such a long history of bait-and-switch is absolutely crazy. (Raising prices repeatedly, lowering rate-limits repeatedly, changing the terms, banning calls which contain "OpenClaw", turning on their IDE partners, turning on their enterprise partners.)

AFAICT anyone who's ever shown faith in Anthropic has been immediately exploited by them to some degree. They will quickly get the reputation of being "the Oracle of AI companies".

I wouldn't even value them at half of OpenAI.

Sam Altman appears to represent a significant liability for OpenAI’s success from this point forward. A big portion of the driver for Anthropic’s meteoric rise over the last six months appears to be folks recognizing “it’s that AI startup not run by Sam Altman.” Anthropic has amazing tech, but its biggest asset at the moment seems to be that “it’s not OpenAI.”

Not saying that’s right or wrong, but it’s clearly a factor holding OpenAI back at this point.

OpenAI’s models could be materially better than Anthropic’s and I still wouldn’t use them because I don’t want to support Altman.

I heard my kids argue last night. My daddy is so tall. My daddy is bigger than the house. No my daddy is bigger than a roller coaster. Yeah? Well my daddy is 50km. You mean he is long? Yes, my daddy is longer than you.

I was cracking up. I'm 5'7 on a good day. I feel like that's how valuation works. We are propping up five foot tall giants.

Ah, it’s a good time to check in with gwern on our conversation about oAI vs Anthropic: https://news.ycombinator.com/item?id=40816755 and our predictions (ca two years ago).

Upshot - poetry expertise does not seem to be the primary focus these days, perhaps to the detriment of the entire world. We did move on from training scaling to “test time” scaling (which I hate as a name btw), Ilya does not seem to have been needed, (although I am really curious what he’s building).

My prediction that you want to be deeply embedded and really rich and part of global infrastructure feels good. My suggestion that oAI / MS would be able to use the lead in 2024 to extend was wrong.

Neither of us talked much about coding as a product that would drive value and behavior, which is super interesting to me, we were probably six months from seeing real competence of any sort there way back in June 2024.

We both seemed to think there would be a single breakout company, or could be one, (although I did suggest buying the basket), clearly not the case with GOOG oAI and Anthropic all posting serious revenues this last quarter / year.

One area of Anthropic that was nascent in 2024, but that I have come to think is super valuable is their mechinterp group. I still don’t see work done by other labs (at least published) to nearly the quality of Anthropic. And the group has clearly moved into a period of productivity; there’s a good chance in my mind it could provide a truly enduring strategic advantage as a tool to be used by the taste makers steering the ship. In 2024, interpretability seemed almost impossible to get a handle on — today, the sustained chipping away at the problem makes a lot more look possible.

Unicorns, strapped with rockets, too busy looking at each other to realise the Earth is far gone.

They'll kill us all, or they'll kill each other. They sure as hell ain't making the world a better place, like they promised.

At this point I think it’s more important to have a solid workflow and understanding of how [insert your favorite model here] works and its capabilities, than chasing the next shinny release jumping back and forth between companies. I just finished my first large project with Codex and it is hard for me to believe Claude can be much better. It may be a bit better or worse, but again, they are all so good now that the user is the one driving the difference.

I guess the competition lost lots of time in focusing on image and video generation. While they're fun gimmicks, I still really haven't seen the value in AI-generated image/video, especially when considering the greater costs involved.

Doubling down on coding was just infinitely smarter. Has there actually been a successful company which uses AI images and video effectively?

Having used both Anthropic and OpenAI models at $work via copilot extensively, I have to say GPT 5.5 currently is best at getting work done with minimal mistakes. However, Claude Code is way ahead of OpenAI Codex in terms of harness features and tooling. MCPs, skills, sub agents, these all were pioneered in Claude Code first. Perhaps that contributed to Anthropic's success.

codex gtp-5.5 is far superior to opus 4.7 working on large projects

They are far far better at marketing than OpenAI

In this game, who wins - in the long term - is who has the best model: so far OpenAI is ahead, so in the long term this is what matters. However, for the same reason, if in the future open weight models will be very near the quality of frontier labs, Anthropic and OpenAI will be out of business very soon. The game they play only make sense if their SOTA models do things that other models can't do at a comparable level.

I dunno, the latest Opus models seems to be tuned to waste money... and Claude is kinda lazy lately?

I've seen fewer people insisting OpenAI has a moat lately, but I'm still not sold the big winner will be either of these two in the end.

qazinform.com seems to be shadow-banned (and posted only by OP): https://news.ycombinator.com/from?site=qazinform.com

UPD https://en.wikipedia.org/wiki/Kazinform

I get the feeling this also means AI works very well for the general coding tasks and that's their biggest success in terms of difficulty AND people paying for it.

Of course every AI company has been over promising and pumping the numbers as much as possible but OpenAI has been hitting the reality wall more because both their people not being able to keep improving at a faster rate and their whole cost structure and financial plates spinning.

This doesn't invalidate the fact Anthropic is also overhyped to the max for their IPO.

Most people think the current valuation is for the models themselves. Actually, they're building the infrastructure for the next 50 years.

How much dilution? Who’s getting the value?

Bernie Madoff would be jealous. Stealing all open source and reselling "git clone" + "sed" for $1 trillion is something he did not achieve.

The chutzpah is remarkable.

my take it's because of the naming: Amodei, Claude and Mythos have this money-throwing vibe to it

Some says the founder worked at Baidu before. Is that true?

All overvalued.

Pointless article (like much of the AI marketing hotness and spin room).

> The new valuation is nearly three times higher than the company’s February valuation, when Anthropic was estimated to be worth around $380 billion.

> In March, OpenAI was valued at $852 billion following a record $122 billion funding round.

Basically, today (Late May) we're declaring Anthropic the most valuable. They've nearly tripled in value since February. But also, OpenAI was $852B in March and presumably has grown since then.

In a few weeks we'll either have a new rounding of funding for OpenAI or they'll announce their IPO and the hype train will be abuzz that they're now the most valuable.

I think claude is much better than chatgpt

I think claude is much better than Chatgpt

This is depressing. Anthropic really is the last company we want to see leading this race, given how greedy they are. Let's not forget all of the lying and gaslighting too. The creator of OpenClaw made this I believe: https://clawd.rip

Stealing peoples tokens because you use a product they don't like... That shows the morals they have. Actions speak louder than words. Disabling peoples caches because they disable telemetry was another juicy one that I don't believe is on this site. In fact there are far more I remember that aren't even listed here.

The headline is false. First off, OpenAI hasn’t raised a recent round so you can’t compare these two companies randomly like this. Second, Anthropic is known to have accounting methods that give it more revenue that they would have if they used the same practices as OpenAI. And neither of these companies are known to be doing gaap accounting.

Investors of both should read this: https://open.substack.com/pub/sublius/p/srt-introspect-why-c...

A sign of a bad developer is they cannot fathom switching from claude cli. In their mind it's claude or nothing. Despite things like codex existings. Opencode existing. Cursor + Compose 2.5 existing.

These are the new .net developers who will know nothing but c# for 20 years.

Bummer, they are the least friendly to open source, and the most incompatible with free use of your subscription via your own tools/custom harnesses.

The models aside, my impression is that Anthropic is winning in large part because of very pragmatic and high-velocity product development on top of them; like with Claude Code.

Like actually iterating hard to make them useful. Many, many details matter here.

I haven't tested the similar OpenAI/Google tools in detail lately though. Previously I found them way too generic and unpolished to be useful.

Is there something to this?

Either they are getting fleeced or they are getting very good terms for the investments

It’s because the programming works.

OpenAI. Spent its resources on AGI whilst Claude worked on making programming work.

Google Gemini is out of the race entirely its programming AI is a joke.

Start what?

ChatGPT dropped the ball for a while that most devs and technical people went to Claude for a year or more, they still probably have the most normie market share + are at least trying to win back some of that delay in their latest model so it'd be interesting to see