[larodi]

One thing I can tell you is you are either favored by Anthropic, or your version of the CLI does not exhaust limits, or there's some major bug, as two people around me (myself included) claim it took half an hour to hit the ceiling. Which makes it practically unusable, where the same workflow a day ago produced a good 5-6 hours of workload with several agents.

[witx]

They are most likely shills from Anthropic, there's quite a few here everytime new models come out.

[miyoji]

That's not fair. Simon is a well-known shill for the entire AI industry, not just Anthropic.

[simonw]

What's your definition of "shill"?

[miyoji]

Merriam-Webster: noun, 1b: one who makes a sales pitch or serves as a promoter

You might want to ask the guy who said it first what he meant; I was just pointing out that your work isn't particularly Anthropic-biased, in my experience.

[Jensson]

Probably means fan, shills have undisclosed ties and I doubt he means Simon has undisclosed ties to the entire AI industry, that would be very impressive if so.

[supern0va]

Ah, yeah. I've noticed people also starting to just use "slop" to also mean "anything I see online that I don't like" now, too.

Words apparently don't mean anything anymore.

[cedws]

It’s not meant for subscription users; the subscriptions are just the gateway drug to Enterprise pricing which Anthropic intends to use to juice their numbers before IPO.

[piokoch]

Monetization is coming. They'll tell companies, AI is replacing your workers, so it is still worth to pay 100K/year for the license, as those AI are not going to jump to other job, get sick, be late, complain, require free coffee and so on.

Soon the times of AI for $20/$200 a month will be long gone.

[tarkin2]

Get people hooked, tell them spending time coding is no longer needed, let their skills deteriorate, tell them they need cough up for a licence to do their job

Forcing developers to pay for models that were build on code they scraped scott-free

A tax to do their job that developers are jumping at the chance to pay

Everybody's finally realising that node dependencies are a threat, but letting these AI companies gatekeep the industry is a bandwagon people are scrambling towards

[witx]

> Forcing developers to pay for models that were build on code they scraped scott-free.

Yes this makes me sad behound explanation. Specially when I see open source developers happily using these tools. These companies stole your, free, hard work and charge you a subscription!! Not to speak about them torrenting books and (most likely) training on private repos.

This and devs paying a subscription to use a tool that is marketed as trying to replace them.

I had 150$ monthly budget thatbI used for various open source projects and I've cut that entirelly.

[simonw]

> These companies stole your, free, hard work and charge you a subscription!!

In case you weren't aware, Anthropic, OpenAI and GitHub Copilot all have programs that provide access to open source maintainers for free:

GitHub: https://docs.github.com/en/copilot/how-tos/copilot-on-github...

Anthropic: https://claude.com/contact-sales/claude-for-oss

OpenAI: https://developers.openai.com/community/codex-for-oss

[andriy_koval]

Was there comprehensive survey amongst maintainers that its fair price for decades of hard work?

[majora2007]

I don't get what you're saying. You're frustrated that Open Source projects were used to build these AIs and that OS devs (or devs in general) are paying to use AI.

Then you say you had money that you used to donate(?) to OS and have cut that because of the frustration?

Open source just means sharing the source code for people to learn off or have the ability to customize on their own. I don't think there is any need to be frustrated about that (now if it was copyright/private of course).

[witx]

> Open source just means sharing the source code for people to learn off or have the ability to customize on their own.

Yes people, not corporations. The point is there a licenses to be respected that weren't.

[lkjdsklf]

Model training pretty clearly falls under fair use.

We could fix that, but it requires a political will to change the law.

[paganel]

> Forcing developers to pay for models that were build on code they scraped scott-free

That's also caused by some very smart (even brilliant) developers (you can see many of them in this very thread) choosing to be oblivious about all this and bury us all under, hoping that they'll be among the last ones to go. Writing this down I realise that they maybe aren't all that smart.

[thewebguyd]

I've been saying this since the beginning, the rug pull is coming. If these models can eventually replace a human worker, there is no reason these companies won't charge (and get away with it) very close to a typical SWE salary.

It would not surprise me one bit to see anywhere from $80k-$100k/seat pricing.

[andriy_koval]

Unless there is competition (e.g. Chinese models, taking you 80% there, but costing 20x less)

[larodi]

As someone noted here recently - use the frontier models as much as u can, while you can.

[dualvariable]

AI for $20/month won't ever go away, but it won't be the absolute latest and greatest frontier model.

Most of us don't need a model that can prove the Riemann hypothesis or Goldbach's conjecture in order to get work done.

[miroljub]

Thankfully, we have Chinese models we can use for a fraction of the price.

Not everyone needs a Ferrari to go for a weekly shopping.

[baq]

A Ferrari will likely lap you when you’re racing, though, and the market and the economy is a race. You’ll be facing a question soon, or your employer will, whether to spend a significant chunk of free cash on fable-class tokens or on literally anything else instead - wages and salaries included.

[iugtmkbdfil834]

<< You’ll be facing a question soon, or your employer will

Maybe? If you talk to executives, the impression that I am getting is that they tend to be somewhat misinformed at best, which, yes, is bound to result in some really bad decisions down the road. But, and it is not a small but, the ones I did talk to ( and, amusingly, those are the ones with strong opinions ) don't seem to have a lot, um, practical exposure to this tech beyond what they heard at the watercooler. Honestly, it is kinda infuriating. And all this before we get to how companies want to say they use AI, but also keep cost down.

[miroljub]

Yeah, sure. In the same way I can see only Ferraris driving as taxis, company cars, transport vehicles, used by post, delivery services ...

You and your work are not that special, you are not participating in car races, and you don't need a Ferrari.

[desmond1303]

Or use API billing? We have access to it at my company with no limits

[simonw]

Are you on the $100/month subscription?

[joshstrange]

I am, and I used up the entire 5 hour window in 8min using the highest thinking setting. It also ate up $15 of extra usage before I noticed.

I’ve done the same thing with opus multiple times with no issue. According to ccusage I racked up just shy of $100 of tokens using Fable.

It spun up subagents or workflows or whatever so obviously that contributed but “double opus” was not my experience. I’ve done the exact same prompt with opus on the highest setting and only once before (not even while using this prompt) hit my limits.

My prompt? I’m not a prompt wizard or anything but it was literally:

> Please review the uncommitted code in this repo for bugs/issues/code smells.

I use variations on that all the time with opus and never had issues. I figured it was a good one to kick the tires with Fable. Little did I know it would mean no more Claude Code for the next 4.5hrs (unless I wanted to pay) after this being the first time I had used CC that day (yesterday).

All in all, a pretty crappy first experience.

[simonw]

Try running this command: and see what it thinks you spent at API prices:

  uvx agentsview usage daily

Then edit the config file to add Fable pricing as described here: https://til.simonwillison.net/llms/agentsview-custom-model-p...

And run the command again. I get $126.89 for yesterday.

[joshstrange]

Hmm, I tried that and made the config file change but it didn't work for me. I just see:

    DATE        INPUT    OUTPUT   CACHE_CR  CACHE_RD   COST     MODELS
    ----        -----    ------   --------  --------   ----     ------
    2026-06-09  142015   85315    321224    6880110    $10.96   claude-fable-5, gpt-5.5, claude-haiku-4-5-20251001

I tried to filter down to just fable (or 5.5 so I could deduct it) but the `--agent` flag doesn't seem to work how I'd expect...

I think the $10.96 is coming from gpt-5.5 since I switched to it once I exhausted all my usage on CC. CCusage reports completely different numbers so I don't know which one of those is right.

Thanks for trying, for yesterday ccusage says "$92.02" for claude, which I assumed was the Fable usage.

[simonw]

If you run this:

  uvx agentsview serve

You'll get a localhost web application which makes it much easier to filter by model.

[joshstrange]

That's very interesting, I had not used agentsview at all before today and I'll have to keep that in my back pocket.

Unfortunately it's not telling the whole story. The last message from the _only_ Fable session it monitored was:

> The data layer looks clean — <REDACTED>. Now waiting on the 11-angle workflow — verification and the gap sweep run after the finders; I'll compile the full ranked findings list when it completes.

And my memory jives with that, I could see in the footer that it had spun up 11 agents (though agentsview says it used 0 subagents, don't know if it was "actually" workflows that it spun up?). It's like it didn't record the sub-sessions/sub-agents info?

I'm still shocked that my prompt (which I now can see thanks to this tool) of:

> Please review all the uncommitted work in this repo and identify any issues.

was able to burn so much, so quickly, and, most frustratingly, without actually doing anything useful because killing it was my only option lest it spend even more of "extra usage".

Overview of usage: https://cs.joshstrange.com/RjGzWVXy

Stats for that 1 session: https://cs.joshstrange.com/Fj5qv1wl

[jsw97]

simonw, if you are not bumping up against the same false-positive guardrail problems and budget consumption that everyone else is, then that is something worth digging into. I would normally say that's crazy but IPOs put weird pressure on companies.

[simonw]

I've had a couple of guardrail blocks.

I've been watching my usage quota bars drop as I use the model, so I don't think I have a weird quota issue going on here.