Measuring Claude 4.7's tokenizer costs • Hacker News Board

Re: Measuring Claude 4.7's tokenizer costs

#151

Post by QuercusMax » Fri, Apr 17, 2026, 6:10 PM UTC

iknowstuff wrote: post #8

Interesting because I already felt like current models spit out too much garbage verbose code that a human would write in a far more terse, beautiful and grokable way

I had a case yesterday where Claude wrote me a series of if/elses in python. I asked it if it could use some newer constructs instead, and it told me that I was on a new enough python version that I could use match/case. Great!

And then it proceeded to rewrite the block with a dict lookup plus if-elses, instead of using match/case. I had to nag it to actually rewrite the code the way it said it would!

Re: Measuring Claude 4.7's tokenizer costs

#152

Post by colechristensen » Fri, Apr 17, 2026, 6:12 PM UTC

rambojohnson wrote: post #149

So intelligence has turned into a utility per Sam Altman et al., and now the same companies get to hike the price of accessing it by 20–30%, right as it’s becoming the backbone of how teams actually ship work. People are pushing out so much, so fast that last week’s output is already a blur. I’ve got colleagues who refuse to go back to writing any of this stuff by hand. And now maintaining that pace means absorbing a…

Like every startup ever, they were selling it to you at a loss to compete for market share and are slowly increasing pricing. Duh.

Re: Measuring Claude 4.7's tokenizer costs

#153

Post by DeathArrow » Fri, Apr 17, 2026, 6:15 PM UTC

beej71 wrote: post #65

News like this always makes me wonder about running my own model, something I've never done. A couple thousand bucks can get you some decent hardware, it looks like, but is it good for coding? What is your all's experience? And if it's not good enough for coding, what kind of money, if any, would make it good enough?

Unless you use H100 or 4x 5090 you won't get a decent output.

The best bang for the buck now is subcribing to token plans from Z.ai (GLM 5.1), MiniMax (MiniMax M2.7) or ALibaba Cloud (Qwen 3.6 Plus)

Running quantized models won't give you results comparable to Opus or GPT.

Re: Measuring Claude 4.7's tokenizer costs

#154

Post by wslh » Fri, Apr 17, 2026, 6:15 PM UTC

rambojohnson wrote: post #149

So intelligence has turned into a utility per Sam Altman et al., and now the same companies get to hike the price of accessing it by 20–30%, right as it’s becoming the backbone of how teams actually ship work. People are pushing out so much, so fast that last week’s output is already a blur. I’ve got colleagues who refuse to go back to writing any of this stuff by hand. And now maintaining that pace means absorbing a…

> So intelligence has turned into a utility.

"Utility" is close, but "energy source" may be closer. When it becomes the thing powering the pace of work itself, raising prices is less about charging for access and more about taxing dependency.

Re: Measuring Claude 4.7's tokenizer costs

#155

Post by minimaxir » Fri, Apr 17, 2026, 6:16 PM UTC

_fat_santa wrote: post #62

A question I've been asking alot lately (really since the release of GPT-5.3) is "do I really need the more powerful model"? I think a big issue with the industry right now is it's constantly chasing higher performing models and that comes at the cost of everything else. What I would love to see in the next few years is all these frontier AI labs go from just trying to create the most powerful model at any cost to ac…

Many people were hoping that Sonnet 4.6 was "Opus 4.5 quality but with Sonnet speed/cost" but unfortunately that didn't pan out.

Re: Measuring Claude 4.7's tokenizer costs

#156

Post by wslh » Fri, Apr 17, 2026, 6:17 PM UTC

colechristensen wrote: post #152

rambojohnson wrote: post #149

So intelligence has turned into a utility per Sam Altman et al., and now the same companies get to hike the price of accessing it by 20–30%, right as it’s becoming the backbone of how teams actually ship work. People are pushing out so much, so fast that last week’s output is already a blur. I’ve got colleagues who refuse to go back to writing any of this stuff by hand. And now maintaining that pace means absorbing a…

Like every startup ever, they were selling it to you at a loss to compete for market share and are slowly increasing pricing. Duh.

But the unique thing about AI is that the "world" is depending on it like water, oil, gas, etc. Not just a specific use case.

Re: Measuring Claude 4.7's tokenizer costs

#157

Post by Aurornis » Fri, Apr 17, 2026, 6:18 PM UTC

louiereederson wrote: post #10

LLMs exist on a logaritmhic performance/cost frontier. It's not really clear whether Opus 4.5+ represent a level shift on this frontier or just inhabits place on that curve which delivers higher performance, but at rapidly diminishing returns to inference cost. To me, it is hard to reject this hypothesis today. The fact that Anthropic is rapidly trying to increase price may betray the fact that their recent lead is a…

> It's not really clear whether Opus 4.5+ represent a level shift on this frontier or just inhabits place on that curve which delivers higher performance, but at rapidly diminishing returns to inference cost.

I think we're reaching the point where more developers need to start right-sizing the model and effort level to the task. It was easy to get comfortable with using the best model at the highest setting for everything for a while, but as the models continue to scale and reasoning token budgets grow, that's no longer a safe default unless you have unlimited budgets.

I welcome the idea of having multiple points on this curve that I can choose from. depending on the task. I'd welcome an option to have an even larger model that I could pull out for complex and important tasks, even if I had to let it run for 60 minutes in the background and made my entire 5-hour token quota disappear in one question.

I know not everyone wants this mental overhead, though. I predict we'll see more attempts at smart routing to different models depending on the task, along with the predictable complaints from everyone when the results are less than predictable.

Re: Measuring Claude 4.7's tokenizer costs

#158

Post by wellthisisgreat » Fri, Apr 17, 2026, 6:18 PM UTC

_pdp_ wrote: post #29

IMHO there is a point where incremental model quality will hit diminishing returns. It is like comparing an 8K display to a 16K display because at normal viewing distance, the difference is imperceptible, but 16K comes at significant premium. The same applies to intelligence. Sure, some users might register a meaningful bump, but if 99% can't tell the difference in their day-to-day work, does it matter? A 20-30% cost…

Does anyone here use 8k display for work? Does it make sense over 4k?

I was always wondering where that breaking point for cost/peformance is for displays. I use 4K 27” and it’s noticeably much better for text than 1440p@27 but no idea if the next/ and final stop is 6k or 8k?

Re: Measuring Claude 4.7's tokenizer costs

#159

Post by solenoid0937 » Fri, Apr 17, 2026, 6:24 PM UTC

cbg0 wrote: post #15

solenoid0937 wrote: post #7

Earlier quoted context omitted.

People complain about a lot of things. Claude has been fine: https://marginlab.ai/trackers/claude-code-historical-perform...

That performance monitor is super easy to game if you cache responses to all the SWE bench questions.

You dramatically overestimate how much time engineers at hypergrowth startups have on their hands

Re: Measuring Claude 4.7's tokenizer costs

#160

Post by Aurornis » Fri, Apr 17, 2026, 6:24 PM UTC

UncleOxidant wrote: post #95

ZeroCool2u wrote: post #71

Earlier quoted context omitted.

Whenever we get the locally runnable 4k models things are going to get really awkward for the big 3 labs. Well at least Google will still have their ad revenue I guess.

Given how little claude usage they've been giving us on the "pro" plan lately, I've started doing more with the various open Qwen3.* models. Both Qwen3-coder-next and Qwen3.5-27b have been giving me good results and their 3.6 models are starting to be released. I think Anthropic may be shooting themselves in the foot here as more people start moving to local models due to costs and/or availability. Are the Qwen model…

I've also been using the Qwen3.5-27B and the new Qwen3.6 locally, both at Q6. I don't agree that they're as good as pre-Opus Claude. I really like how much they can do on my local hardware, but we have a long way to go before we reach parity with even the pre-Opus Claude in my opinion.