[matheusmoreira]

> As a non-US citizen Opus 4.8 is the best American LLM I will ever have access to.

> Each month that I pay Anthropic is now a depreciating value -- I'm paying for models I'll never be able to access, while other models are able to catch up.

Excellent point... That made me rethink my payments to Anthropic. As one of the foreign peasants who was banned from accessing Fable by the land of the free, it's become really hard to justify giving Anthropic any more money. I'm very tempted to switch to GLM 5.2.

[nozzlegear]

> banned from accessing Fable by the land of the free

It should go without saying, but "the free" in "the land of the free" refers to Americans, not everyone else. Not sure if you were trying to make it sound like a great irony that "the land of the free" would exclude or ban people, but if so it doesn't quite hit the mark. A more cutting criticism would be that the land of the free isn't letting one of its own companies freely compete.

[matheusmoreira]

Fair point. The fact that even americans got cut off from Fable is definitely ironic though. I suppose that's going to be fixed now that they're implementing identity verification. And even that runs smack into the concept of freedom since freedom under surveillance isn't real.

[inlined]

But given the hoopla about how a driver’s license isn’t valid in the SAVE act because it doesn’t prove you’re a US citizen, how is drivers license verification going to satisfy the US national requirement here?

[cryptonector]

It probably won't. It will probably have to be a U.S. passport.

[yurishimo]

Many states have extra markings on the card for citizens vs legal residents. A valid birth certificate would also be enough since the US is one of the only nations to recognize birthright citizenship.

Also worth pointing out that most Americans don't have passports and getting one can take anywhere from 6 weeks to 6 months.

Also, you'd have to be smoking something to assume that the three letter agencies are not going to be combing through Anthropic's customer list and assessing for fraud and foreign bad-actors. If you lie, they'll probably catch you eventually and most people are not stupid enough to lie about their identity to the US government and think they'll get away with it forever.

[mcv]

It doesn't even refer to all Americans. Just the wealthy ones. And even that is increasingly reduced to just the wealthy ones who support Donald Trump.

[tommit]

It's always been a catchy propaganda slogan. Nothing more.

[3stacks]

What is stopping you from switching to GLM 5.2 now? Have you tried it out yet?

I'm playing with Deepseek a lot more via OpenRouter recently and the only major downside I can see is the usage billing over the monthly plan

[matheusmoreira]

Existing setup. I'm already used to Claude Code I guess. I actually spent time and tokens customizing and fixing it. Have a patcher utility that modifies the binary in order to disable telemetry and remove performance reducing language in the system prompts. Every update I spend some Claude tokens dissecting the newest executable and integrating it with my patcher.

The monthly subscription is also a major hurdle for me. The "high end frontier models for low prices" aspect is a major reason. I think I'm getting a lot of value from my subscription, given that the API prices are like a hundred times higher.

However, there's also a psychological factor here. These subscriptions are like the gachas of the software world. I got "addicted" to them. I developed workflows around achieving 100% weekly usage. Sometimes Anthropic randomly resets weekly usage and I scramble to get the most out of it. I'll point Claude at things and then just have it run hundreds of code review agents. I ran out of projects to do this on and started doing it to my favorite open source projects instead, looking for things to contribute.

I think with usage-based pricing I wouldn't use LLMs quite so freely. It'd probably cure my "addiction" too, but the problem is I'm not sure whether that's a good thing, since this "addiction" has been a somewhat positive force in my life. It's driven me to start new projects and also make major progress on existing ones. It brought me out of a slump. I'm a little afraid of moving away from Claude and not being as driven as I was before.

[eturkes1]

Are you me? Anyway I wonder if you've played with the following, they've been what I've been using to patch CC:

https://github.com/skrabe/tweakcc-fixed https://github.com/skrabe/lobotomized-claude-code

[matheusmoreira]

Yeah I know about tweakcc, even had some custom prompts committed to my dotfiles repository for a while. For some reason it wasn't working for me, don't remember why.

At first I used this shell script here:

https://gist.github.com/roman01la/483d1db15043018096ac3babf5...

Someone posted it on HN on the Opus degradation thread:

https://news.ycombinator.com/item?id=47660925

I just kept using it to fix the prompts for a long time. Eventually I asked Claude to port it to Python to make it easier to maintain. Now every month or so Claude dissects the Claude Code binary and integrates the script with the newest version.

https://github.com/matheusmoreira/.files/blob/master/%7E/.lo...

[eturkes1]

Thanks for the info. Yeah tweakcc is finicky, in fact it broke again for me with the latest CC update. Maybe I'll end up with my own solution as well in the end.

[Scoundreller]

Meanwhile, as an irregular user of heavy AI, I like the usage-based billing of deepseek. Made a bunch of optimizations to my vibe-coded codebase, created some extra modules, using the chat... has costed me 33 cents so far in a couple weeks.

[djeastm]

Yeah for a side project I just switched over to DS4 and between Pro and Flash a few days ago I've spent $1.96 both for code work and some large LLM text processing tasks.

It feels to me as good as the last version of Opus I was using (I think 4.6 or 7?). I'm going direct through the DeepSeek API, not Openrouter.

[matheusmoreira]

33 cents? That's insane... I set up an API cost meter on my Claude Code status line and it frequently runs into the hundreds, sometimes thousands of dollars.

[Scoundreller]

My build/work is probably a lot simpler than yours, but still, a huge value for me for just 33 cents.

Minimum spend/fill is US$2 plus Chinese vat (dunno why but not gonna complain about 6%)

[patates]

For the 100$ I'm paying for Claude, I'm pretty sure I can use Deepseek way more than I can use Opus via the plan.

[3stacks]

Fair. If you trust the stats on Codeburn I've used $6k USD of tokens in the last month with Claude as a very heavy user, so our situations may differ

[zubairov]

The downside are opportunity costs - not using other models to make better decision, I guess, right?

[Amekedl]

usage billing over the monthly plan when deepseek is over x25 cheaper?

[3stacks]

Maybe I'll try out a month on the usage plan and see how it compares dollar for dollar. I think I'm squeezing about as much as I can out of Anthropic under the plan

[digitaltrees]

As an American founder I am sorry for this. I think the world needs to band together to take geopolitical risk out of AI.