Agreed. I think the Chinese labs are proving that OpenAI and Anthropic don't have a moat in almost every aspect, especially pricing. I also think people are getting annoyed with the constant lift and shift. I've seen more folks drop Claude Code and Codex, specifically, because of the lock-in it provides the providers. I'm curious to see how people standardize on tooling adjacent and if Anthropic, Google or OAI move to block utilization akin to the games Anthropic has been playing as of late.
I think the end game is routed model usage and SLMs. I think Apple is going to prove this in the consumer space pretty handily and I'm curious how the Android ecosystem responds since the hardware is considerably lacking in model performance. I think Apple has a huge opportunity here, as much as I don't like their current ecosystem of walled garden. They did position themselves very well with ARM and custom chips for their hardware. Hopefully the broader ecosystem of ARM and Linux are able to make some headway and we see a more formalized, and broadly accepted, architecture to capitalize on.
is there an alternative to codex that “just works”? by just works i mean i can install as an app in 1 minute, and i get web search, skills, mcp servers, etc? Bonus points if it can control my chrome tabs like codex can, and if it offers remote control from my iPhone (chatgpt app) so i can kick off tasks while i’m out for a walk. Even more bonus points if i can, with 1 button click, share my chats or share the results of a session as a “site” (vercel style).
I’m sure you could put something similar together with a bunch of duct tape and 2 weeks of effort, but it won’t work nearly as nicely nor out of the box. so…what am i missing?
My company has an agreement with the big providers and while i'm pretty sure they think about how to get budget back, its an competitive advantage and normal people will not learn different model behaviours.
going to IPO is a sign of confidence , you need to report a lot of things, that private companies don't. This is an exact reason chinese labs do not rush to go public. They wish to go , but money flow that is not as good.
On the same note. if spacex is doing datacenters on earth successfully what's wrong with that? They rented cloud infra to a #2 or #3 provider in the world after < 2 years in business. It's a success, no?
> if spacex is doing datacenters on earth successfully what's wrong with that? They rented cloud infra to a #2 or #3 provider in the world after < 2 years in business. It's a success, no?
If you get hired as a staff engineer and do the work of a junior, what's wrong with that?
Clearly xAI (now part of spaceX) did not raise funds to be a data center. The margins are way different. There are plenty of recent IPOs in that area that are worth at most billions not trillions.
> going to IPO is a sign of confidence , you need to report a lot of things, that private companies don't.
This isn't going to IPO. This is rushing to IPO. It is a sign of confidence that the market or wider environment might crash soon so we need the liquidity now.
> This is an exact reason chinese labs do not rush to go public.
Maybe or maybe not. If you are referring to Chinese labs - both the Hong Kong and China stock market are way weaker than Nasdaq. It's not comparable. Check all the recent Hong Kong IPOs that have tanked.
So no, reason not to might just be: no money in it.
You’re not gonna get nuanced discussion on spacex or anything Elon related here these days. Most of this site is Reddit lite at this point including their milquetoast progressive opinions (Elon bad being one of them).
I don't think anyone has a firm grasp on actual inference costs -- including the research and training that has gone into those models. We've got near-frontier capabilities from open source models from China at pennies on the dollar compared to US big tech rollouts. OpenAI and Anthropic are heavily subsidizing their inference -- no wait, they are charging the most they can get away with before going public. Where is the truth?
> I don't think anyone has a firm grasp on actual inference costs.
There are huge numbers of users (myself included) that do have an exact idea of what inference costs are - on open models. We can buy tokens from 3rd parties that have no motivation to subsidize our use. That's to say, there's a fair marketplace[1] and we're hanging out there.
If you want to say "I don't think anyone has a firm grasp on actual inference costs on these proprietary/closed models", then I could agree with that.
There is no way I'm believing DeepSeek can charge less than $1 USD for their pro model while Opus costs over 25x more, yet their price is less than the cost of running it?
It would seem strange, if they were operating in the same economy, but they don't. DeepSeek operates in an economy with a high degree of central planning.
China subsidizes strategic industries, and they have heavily done so with AI. And DeepSeek specifically has said they have no commercialization plans.
DeepSeek is not the only provider of inference for their models. Chinese subsidies likely do explain DeepSeek's ability to provide inference cheaper than other providers, but even a US provider like DeepInfra can serve DeepSeek 4 Pro at $1.30/M in and $2.60/M out. Unless American labs are doing something wildly inefficient, it feels safe to assume Anthropic has some profit margin on inference at API prices.
They may, neglecting overhead R&D. But also, some suspect that US models are significantly heavier than DeepSeek in resource consumption by multiple measures
It’s generally established that Anthropic/OpenAI are going for all out performance with big VC dollars at the expense of efficiency and China has geopolitically limited compute and an inventive to compete on value per dollar.
> OpenAI and Anthropic are heavily subsidizing their inference -- no wait, they are charging the most they can get away with before going public. Where is the truth?
Both. They are charging the most they can get away with and that amount is still heavily subsidized by VC capital.
> I don't think anyone has a firm grasp on actual inference costs -- including the research and training that has gone into those models
We know roughly how much these companies spend and what their revenues are. Based on that, they'd have to more than double revenue (without spending more money) just to stay even, and that's not good enough given how deep in the hole they are.
> OpenAI and Anthropic are heavily subsidizing their inference -- no wait, they are charging the most they can get away with before going public. Where is the truth?
Both are true. I mean, I'd be willing to spend a bit more than I do now, but not more than double, and neither are most companies. The company I work for is currently investigating how to reduce LLM spend, not looking to spend more.
We have a firm grasp on actual inference costs from the various open weights model providers on OpenRouter. They don't have the money to subsidize inference and it's quite a competitive market, so the prices are representative of the costs.
Just outta curiosity, as I’ve never gotten a spend anywhere near that, what variant were you using? Like max context window and fast mode? Or was it just chugging along non stop for three days?
Fast mode max content window. The task was: replace all 1600+ queries from one database to another and make the whole integration test pass. We did multiple passes, with different concerns when changing from database to another. My OpenCode session right now says $4,365.02.
I haven't gotten close to this either before, but now we wanted to move fast because this branch gets conflicts all the time and we want to get over with the migration asap.
It's a bit of a left field question, but I am curious: Let's say that if the company wasn't paying the whole bill but only subsidizing it - e.g, if it paid 90% of the $4000. What would you do?
I don't know, why would I pay to do my job? It's not my first database switch for a startup. Only this time it doesn't take two months of grueling work. I know exactly how this is done, but the amount of grunt programming and testing and repetitive work is just not great. And it's not a task that brings new customers or a new product. Just a mandatory and annoying thing to deal with when we are growing.
And don't get me wrong. Opus did an absolutely horrible job at first, second and third round in this task. You really needed to steer it to get to the right solution.
And now Fable is out. And its first round of code reviews for this huge PR was definitely worth the money too...
Don't think that I'm just shrugging to that number. I see it every day, and I don't like that it's in the thousands now. But for people paying the 100 or 200 dollar plans, I'm not super sure if you will be able to use them in the future if the token price is in the thousands for a bit bigger task...
If I'd pay this from my own pocket, I'd definitely go with DeepSeek or local models and figure it out how to make the best use of them.
> If I'd pay this from my own pocket, I'd definitely go with DeepSeek or local models and figure it out how to make the best use of them.
IOW, you don't really think the value of this work is really worth $4k.
> why would I pay to do my job?
The question is: how long do you think that you employer will be willing to pay for you and Anthropic, if you yourself said if it were your money you'd put some time and effort to work with an open model?
> The question is: how long do you think that you employer will be willing to pay for you and Anthropic, if you yourself said if it were your money you'd put some time and effort to work with an open model?
I wonder what this question really means? Anthropic is useless if you don't know what to do with it. It's very useful if you do, and you can guide it to do the right things. Yes, it will for sure reduce the amount of people we need to hire. But we are always looking for hires who know what they do and can utilize agents to be faster.
But if you think about how long employer is willing to pay 10-20k per month per seat for Anthropic? I can't see this to be feasible and it will have to end at some point.
Regardless of the actual value produced by the models, if I am the CTO of any company that has the budget to spend $10k/month/seat on Claude, I'd take 5%-10% of that to build an alternative in-house.
I'm with you here. We can't slide into a situation where you put a sizable amount of your budget for an American mega corporation if you want to survive in the competition. We need local models and we need them to be good enough to help us.
regardless of whether that's true or not, US companies doing hosted inference of the models coming out of China are also significantly cheaper than those from OpenAI or Anthropic
Just a personal anecdote but I have not hit any more thresholds or limits since switching to the MAX plan and so far, it's been worth it. But I do wonder how long even this will last...
I think subscription models are sustainable, but longer term, we should probably expect to see more prompt optimization happening in the providers inference pipeline. For example, unless you explicitly tell the agent or API to use a specific model, fronting the inference layer with a caching prompt classifier to determine which model to use, and automatically select the lowest cost model would probably already save alot of money (IDK if Claude/OpenAI do this on the backend, but several services I have worked on do some things like this to reduce costs of delivery customer facing inference at scale).
> fronting the inference layer with a caching prompt classifier to determine which model to use, and automatically select the lowest cost model would probably already save alot of money
Unfortunately, that doesn't work within a single session. The K-V cache of a model is intertwined with the model's configuration. Switching models invalidates the cache, meaning everything up to the point of the switchover is processed like a new, uncached input token.
Per Anthropic's pricing doc, an Opus 4.8 cache hit costs 50¢/MTok, while Haiku costs $1/MTok for uncached input.
Model selection works best if sessions are short and self-contained, particularly if the first few interactions can reliably classify the model need. That probably covers most 'support chatbot' use-cases, but it doesn't describe the kinds of heavy agentic automation that really chews through token budgets.
There is a definite financial incentive for people smarter than me to solve the problem, and I don't generally bet against businesses finding ways to reduce costs :)
> The K-V cache of a model is intertwined with the model's configuration.
I don't think this is true if you simply quantize the model or run it with fewer active experts? The underlying weights would stay the same. You could also play further tricks with skipping some of the model's middle layers outright, which works surprisingly well due to how skip connections are used.
I tried ultracode today on the max pro plan. An hour and a half in was all I lasted. Giant review on an entire six month old code base. It found 61 bugs, about ten were notable. Pretty impressed.
Ultracode destroys your limits and I have not found it to be worth it in the slightest, just fyi. I haven’t found any improvement over a local Claude code instance set to opus max.
I have the $100 plan and had almost never run out of credits until I started using the ultracode / workstreams feature w/Opus 4.8..at which point I managed to blow the full 6 hour allocation in like 20 minutes, or so. In fairness, it did some amazing things with the extracted information, but it also strongly suggested that I'd need the $200 subscription *plus* a budget for extra usage.
Instead pay for 3 Chinese models. No max out ever then. I pay for kimi, DeepSeek and Claude. Whenever Claude decides it's over, I can safely continue on very cheap plans.
My bet is they'll keep subsidizing for a considerable period of time, at least 1-2 decades more.
Most AI companies are just testing the waters with paid tiers right now, their greatest fear with increased pricing is folks reverting back to wikipedia, stack-overflow and other public domain organic activity buzzing back to life; that will kill any RoI potential in LLMs forever. They're playing the wait game instead, observing how the digital sphere reacts to every little increase in price.
If that weren't the case, they'd be pricing at lucrative premiums already and even gotten away in short-term considering the increased dependency in the enterprise world. But that'd be like killing for the golden egg too soon and losing all long-term potential.
Once the folks are so addicted to LLMs that even writing a hello world program sounds like a nightmare and coming up with an article draft feels like reinventing Egyptian glyphs, that's when the real pricing hammer will come.
Anthropic and OpenAI won't be around in 1-2 decades if this is their long term plan. People are not going to revert, but go elsewhere. China is proving that it can be done cheaper.
Oh for sure. I've been hopping around from provider to provider for the last few years just depending on who has the most capable / subsidized plans at the moment. I definitely expect there will be a squeeze on subscription costs all around the industry post IPO.
> Nothing is subsidized. Subscriptions are profitable for both Anthropic and OpenAI.
Even if subscriptions are locally profitable (i. e., the cost of the subscription covers the cost of inference), they're still subsidized because they don't cover training and running the company; otherwise, these companies would be profitable.
I can see that being true, and it very likely is true. But isn't infinite VC money and no incentives to optimize operations the reason behind that?
Take a look at China for example - they have no access to NVIDIA, so they're trying to build their own hardware, they have no unlimited funding, so they try to optimize things.
And Anthropic is complete opposite of that - if NVIDIA were to triple their prices tomorrow, Anthropic would still pay them.
In the end, either we all somehow go mad and start paying Anthropic tens of thousands of dollars per month so support this madness, or we will go with whoever isn't lighting cash on fire.
> Take a look at China for example - they have no access to NVIDIA
Not true. Stop following US media spam if needed.
1. Very recently, the US did close a loophole on sanctions that allowed Chinese companies to use NVIDIA hardware outside of China i.e. before that was closed they all had access. The trick was train outside, do adjustments, ship the disks back and use non-NVIDIA in China, but at least the training and endpoints not hosted in China could all use NVIDIA.
2. There's been plenty of reports including fines and bans e.g. to Supermicro on smuggling NVIDIA hardware to China. I doubt it has been stopped. You can't catch everyone.
"Nothing is subsidized" is a wild take. They might be making money on some users, perhaps even most users, but certainly not all. Also, "subsidized" doesn't just mean on compute.
I do, and it's called DeepSeek's pricing table. At the same time, "subscriptions are subsidized" cohort have no data whatsoever, and yet they're in every thread.
Granted, it could still mean that Anthropic just chooses to lose money - but that's Anthropic's choice.
DeepSeek has proven that inference can be much, much cheaper than what Anthropic advertises on their API rates page.
How do you defend these slimey companies? They’re actively running a mob casino and you still have people acting like government is the bad guys here. That doesn’t mean there can’t be better regulation of other markets, but comparing prediction markets to stock markets is a huge stretch.
Disagree, I find their product valuable and use them daily as a source of unusually high quality predictions. When used for this purpose insider trading is a feature that improves the quality of predictions. I see some fraud as in any market, but the overwhelming majority of transactions are voluntary, open and relatively informed within a highly transparent system.
I think that self fulfilling prophecy attempts by deep pockets trying to sway markets by bucking trends generally transfers money from more to less foolish bettors.
A thought experiment: how would you feel about betting on a market that is an the outcome of a medical procedure? On a negative outcome? On a market for a negative outcome of your own procedure?
If the only person who can get the money is you (or your partner or children or whatever), it’s fine as a form of compensation for potential damages.
If anyone, including your surgeon, can take that life insurance policy based on your life, things can go bad pretty quickly (hint: what happens if a profit-maximizing surgeon would earn a lot more money from your policy than from his regular job?).
If it is someone else's? Bad, because I'll just take a life insurance on them and then promise the doctor half of the proceeds if they ensure that the outcome of the procedure leads to an insurance payout.
You lived all your life without these evil companies. Life will go on when they are banished. I don't think you will miss "unusually high quality predictions" after a week.
What predictions? Why is it useful to know what the odds are for Trump to the word “postage stamp” in a specific speech?
Why are the sports odds useful? Word mention market and sports market are the majority of bets after all. Seems like >90% of wagers are useless noise.
Name 7 recent useful ones you actioned based on, one for each day of the last week. I’m very curious what those may be that you use it daily.
When I looked a the site and checked out a few non sport/word wagers, the actual bets were pretty unhelpful because while their summary sounded potentially informative the actual fine print showed that a weirdly constrained timeline of a specific thing was the actual deciding factor, making them useless.
You can look at who is likely to become the next president for USA etc, it helps a lot to see what people who spent some effort looking into it thinks.
Bad example. Americans can only very recently even place trades on PM and even then its through other companies and most people don't know about this. Those markets are often very thin and manipulated by the campaigns themselves.
A better example is wagers on things like when the Iran war will end. That's actually useful 'wisdom of the crowds', and any insider trading is already illegal. Maybe someday those markets would get big enough to allow companies to hedge risks. That's an actual useful application of PMs.
The search function in Jira has always been unusable. It’s perhaps the worst part of the entire platform, but nice to see they’re still focused on adding features I will never use.
I've always thought I was the only one experiencing this and felt like I was crazy.
I guess it's "good" to know that I'm not alone.
The amount of times I've searched for a ticket that I know it's there (because I either have it opened in a different tab, or because I just created it), but can't find, it's just way to many.
The results usually seem completely random to me. It's like the feature never made it out of proof of concept territory. The only advantage of all the email noise Jira sends out is that I can usually search my email for what I'm looking for.
I always got sad when I create a ticket and I see the "ticket created" toast, and then I'm like "oh shoot I forgot to add a screenshot" and go to click the toast to go to my ticket but the toast disappeared. Because then I know that I'm gonna waste the next five minutes of my life looking for it.
FWIW Github has similar shitty search interface. Not sure why.
YouTrack's search is one of the main reasons I use it. Nice query language to filter down on any fields, including custom fields, never had an issue finding things. It's great. With the number of useless search functions in so many products, I'm happy that at least my issue tracking does it right.
That’s maybe the reason today to not build more but not the reason it is bad. America ignored rail for decades in favor of highway systems and now the cost is almost always considered infeasible. We will redo our roads every 5-10 years though.
If it was invested in 50 years ago or more we would be in a different place for sure.
I also live in Chicago and wouldn’t mind walking extra to another stop, but Chicago also has a massive traffic problem, particularly post pandemic. During rush hour, the bus is stop and go already.
I’m really curious how this would pan out here, but it can’t be the only solution.
I think the only way to solve this is to invest much more into making buses nicer & increasing the numbers, and then instituting bus-only lanes on major arterial roads so that taking the bus becomes faster than fighting traffic.
Bus only lanes exist, but not everywhere (some streets are not wide enough). Additionally, at the moment (and really always) there's a ton of construction on the bridges going in and out of the city, causing buses to miss out on the bus-only lanes for years or more at a time. Bus systems are a complicated beast.
You don't have to allow cars on every single paved surface. Cars aren't supposed to be on wide sidewalks, you don't have to allow them on every single road either.
Yeah if they aren't enforcing the bus-only aspect then it's not really a bus-only lane. If a bus only lane exists but people violate it in cars with impunity then obviously it's not going to work.
Wow, this is a weird a comment. Who are "they"? You sound like you think there's some giant conspiracy against JS frameworks. Is the Illuminati behind this? I kid, but a browser feature is kind of what it is. It can take years for features to make it into enough browsers to make them usable. It's quite a bit different than the fluidness of a JS framework.
This discussion comes up all the time and I always have the same response: not everyone needs a full-on framework for what they're doing. They also may need to share that code with other teams using other frameworks or even third parties. The post even mentions that web components may not be a good fit for you.
> Who are "they"? You sound like you think there's some giant conspiracy against JS frameworks.
Yes. There is. The main developers and proselityzers were completely insanely biased against web frameworks (especially React).
It wasn't even a conspiracy. All you had to do was to follow Alex Russel (the person who introduced the idea of web components in the first place) and see his interactions with framework authors and his views towards web frameworks.
The new people in the space driving the specs are hardly any better. E.g. their reactions to Ryan Carniato's rather mild criticism of Web Components is just filled with vile, bile, and hate.
They literally refuse to even admit they have a problem, or want to look at any other solutions than the ones they cook up.
> but a browser feature is kind of what it is. It can take years for features to make it into enough browsers to make them usable.
Strange, browsers push dozens of specs for web components without ever taking any time to see if the yet another half-baked "solution" is actually workable.
Some links to examples of the sort of behaviour you're describing would be really helpful here (I say this as someone who is sympathetic - I work with on a web component/Lit codebase in my 9-5 and I'm not a fan, compared to the React workflow I had in a past life).
Unfortunately, as most of them left Twitter they also temoved all their accounts, so you can only see responses from framework authors like Rich Harris.
Web components are a trend? I've been using them for close to 10 years and they're still not anywhere close to mainstream. Loudly as possible? They've quietly just kind of been there for years.
I think we have a generation of developers that only know React and they're so engrained with it they simply cannot imagine a world without it. If you really can't find a use case for web components then you're living in a bubble.
We have been through all this before with jQuery. The generation of JavaScript developers at the beginning on React only knew jQuery and they really wanted to shoehorn all the jQuery nonsense into the standards. From their perspective it makes complete sense because that is the only one way to do things. They got querySelectors into the DOM.
Now we are seeing the exact same thing again. People only know React, so they want the standards to look like the only one thing they know. That doesn't make it a good idea. Every time this comes up we exchange simplicity and performance for easiness and temporary emotional comfort. Its only a temporary win until the next generational trend comes along.
> If you really can't find a use case for web components then you're living in a bubble.
There's a very tiny use-case for web components. And even there it's riddled with a huge amount of potential (and actual) footguns that "in the bubble" devs have been talking about for a decade at this point, and some which were finally acknowledged: https://w3c.github.io/webcomponents-cg/2022.html (no updates since)
> There's a very tiny use-case for web components.
That's weird, we've been using them at my company for a number of years and there's plenty of other examples of them being adopted elsewhere too. This continues to read as, "it's not React, so it's bad."
There are companies that still use jQuery, or Angular, or Ember, or vanilla JS, or Blazor, or...
Just because tech exists and is usable doesn't mean it doesn't have a list of issues and footguns a mile long, some of which are explicitly recognized by the Web Components Working Group. And which still need 20+ various specs to paper over.
The problem is your comment is missing really important information. You said you returned in 2017, but only mentioned that it was good in 2005. This leaves a 12 year period where it could have declined. You didn't say it was good in 2005 and you already started to see the decline before you left.
You have no more evidence that it declined before the acquisition than after it. It reads as some weird defense of Bezos and then you doubled down by saying management wasn't happy with employee backlash.
Anyways, I agree that the decline did start before the acquisition, like it started for all newspapers. The Internet killed the newspaper. Bezos was supposed to save it.
> It's also the case that The Washington Post brought itself down. I grew up reading WaPo and when I moved back to DC as an adult c. 2017 I got a subscription.
This doesn't really add up given Bezos purchased it in October 2013.
> It also probably did not inspire very much good will from management/ownership when the company's employees started regularly leaking proceedings at company meetings and reporters started making a practice of using social media to criticize management during work hours.
Your thinking is completely backwards. This isn't the first case of a wealthy individual buying journalism in order to destroy it. Why do you think employee backlash happened in the first place?
reply