Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Token Counter, now with model comparisons

simonwillison.net

Post reply on HN
Print view View original

11–20 of 93 posts

Re: Claude Token Counter, now with model comparisons

#11

Post by anonymoushn »

kouteiheika wrote: post #7

> Opus 4.7 tokenizer used 1.46x the number of tokens as Opus 4.6 Interesting. Unfortunately Anthropic doesn't actually share their tokenizer, but my educated guess is that they might have made the tokenizer more semantically aware to make the model perform better. What do I mean by that? Let me give you an example. (This isn't necessarily what they did exactly; just illustrating the idea.) Let's take the gpt-oss-120b…

their old tokenizer performed some space collapsing that allowed them to use the same token id for a word with and without the leading space (in cases where the context usually implies a space and one is not present, a "no space" symbol is used).

Re: Claude Token Counter, now with model comparisons

#12

Post by fooker »

kouteiheika wrote: post #7

> Opus 4.7 tokenizer used 1.46x the number of tokens as Opus 4.6 Interesting. Unfortunately Anthropic doesn't actually share their tokenizer, but my educated guess is that they might have made the tokenizer more semantically aware to make the model perform better. What do I mean by that? Let me give you an example. (This isn't necessarily what they did exactly; just illustrating the idea.) Let's take the gpt-oss-120b…

This is how language models have worked since their inception, and has been steadily improved since about 2018.

See embedding models.

> they removed the tokenizer altogether

This is an active research topic, no real solution in sight yet.

Re: Claude Token Counter, now with model comparisons

#13

Post by aliljet »

This is the rugpull that is starting to push me to reconsider my use of Claude subscriptions. The "free ride" part of this being funded as a loss leader is coming to a close. While we break away from Claude, my hope is that I can continue to send simple problems to very smart local llms (qwen 3.6, I see you) and reserve Claude for purely extreme problems appropriate for it's extreme price.

Re: Claude Token Counter, now with model comparisons

#14

Post by londons_explore »

aliljet wrote: post #13

This is the rugpull that is starting to push me to reconsider my use of Claude subscriptions. The "free ride" part of this being funded as a loss leader is coming to a close. While we break away from Claude, my hope is that I can continue to send simple problems to very smart local llms (qwen 3.6, I see you) and reserve Claude for purely extreme problems appropriate for it's extreme price.

I think an LLM that is a decent chunk smarter/better than other LLM's ought to be able to charge a premium perhaps 10x or 100x it's competitors.

See for example the price difference between taking a taxi and taking the bus, or between hiring a real lawyer Vs your friend at the bar who will give his uninformed opinion for a beer.

Re: Claude Token Counter, now with model comparisons

#15

Post by DeathArrow »

aliljet wrote: post #13

This is the rugpull that is starting to push me to reconsider my use of Claude subscriptions. The "free ride" part of this being funded as a loss leader is coming to a close. While we break away from Claude, my hope is that I can continue to send simple problems to very smart local llms (qwen 3.6, I see you) and reserve Claude for purely extreme problems appropriate for it's extreme price.

Quality of answers from quantized models is noticeable worse than using the full model.

You'll be better using Qwen 3.6 Plus through Alibaba coding plan.

Re: Claude Token Counter, now with model comparisons

#17

Post by dannyw »

kouteiheika wrote: post #7

> Opus 4.7 tokenizer used 1.46x the number of tokens as Opus 4.6 Interesting. Unfortunately Anthropic doesn't actually share their tokenizer, but my educated guess is that they might have made the tokenizer more semantically aware to make the model perform better. What do I mean by that? Let me give you an example. (This isn't necessarily what they did exactly; just illustrating the idea.) Let's take the gpt-oss-120b…

LLMs are explicitly designed to handle, and also possibly 'learn' from different tokens encoding similar information. I found this video from 3blue1brown very informative: https://www.youtube.com/watch?v=wjZofJX0v4M

Also, think about how a LLM would handle different languages.

Re: Claude Token Counter, now with model comparisons

#18

Post by friendzis »

kouteiheika wrote: post #7

> Opus 4.7 tokenizer used 1.46x the number of tokens as Opus 4.6 Interesting. Unfortunately Anthropic doesn't actually share their tokenizer, but my educated guess is that they might have made the tokenizer more semantically aware to make the model perform better. What do I mean by that? Let me give you an example. (This isn't necessarily what they did exactly; just illustrating the idea.) Let's take the gpt-oss-120b…

This is such a superficial, English-centric take, but it might as well be true. It seems to me that in non-english languages the models, especially chatgpt, have suffered in the declension department and output words in cases that do not fit the context.

I have just ran an experiment: I have taken a word and asked models (chatgpt, gemini and claude) to explode it into parts. The caveat is that it could either be root + suffix + ending or root + ending. None of them realized this duality and have taken one possible interpretation.

Any such approach to tokenizing assumes context free (-ish) grammar, which is just not the case with natural languages. "I saw her duck" (and other famous examples) is not uniquely tokenizable without a broader context, so either the tokenizer has to be a model itself or the model has to collapse the meaning space.

Re: Claude Token Counter, now with model comparisons

#19

Post by ipieter »

kouteiheika wrote: post #7

> Opus 4.7 tokenizer used 1.46x the number of tokens as Opus 4.6 Interesting. Unfortunately Anthropic doesn't actually share their tokenizer, but my educated guess is that they might have made the tokenizer more semantically aware to make the model perform better. What do I mean by that? Let me give you an example. (This isn't necessarily what they did exactly; just illustrating the idea.) Let's take the gpt-oss-120b…

There is currently very little evidence that morphological tokenizers help model performance [1]. For languages like German (where words get glued together) there is a bit more evidence (eg a paper I worked on [2]), but overall I start to suspect the bitter lesson is also true for tokenization.

[1] https://arxiv.org/pdf/2507.06378

[2] https://pieter.ai/bpe-knockout/

Post reply on HN