frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

SpaceX files for IPO, targets $1.75T valuation

https://arstechnica.com/space/2026/04/spacex-finally-files-for-ipo-targets-1-75-trillion-valuation/
1•AndrewDucker•1m ago•0 comments

NASA Teams Readying Artemis II Moon Rocket for Launch

https://www.nasa.gov/blogs/missions/2026/03/31/nasa-teams-readying-artemis-ii-moon-rocket-for-lau...
1•Alupis•1m ago•0 comments

Affected by This LastPass Breach? How to Get a Cut of the $24.5M Settlement

https://www.pcmag.com/news/affected-by-this-lastpass-breach-how-to-get-a-cut-of-the-245m-settlement
2•fraXis•2m ago•0 comments

Show HN: Modern AI assisted goals and performance management

https://prfrm.architectfwd.com
1•quintes•3m ago•0 comments

pg_plan_alternatives: Tracing PostgreSQL Query Plan Alternatives Using eBPF

https://jnidzwetzki.github.io/2026/03/04/pg-plan-alternatives.html
1•tanelpoder•4m ago•0 comments

We scanned 73 open-source MCP servers. Here's what source code analysis found

1•sigildev•6m ago•0 comments

PixelPaint – A MacPaint tribute for Apple's 50th

https://blakecrosley.com/paint
1•blakec•6m ago•1 comments

"We enabled the availability of iOS 18.7.7 for more devices"

https://support.apple.com/en-us/126793
4•tech234a•8m ago•0 comments

Describe a dashboard, get a working app in 3 minutes

https://bubbling.dev
1•georgesmith9914•9m ago•0 comments

Iranian president has released an open letter to the American people

https://www.presstv.ir/Detail/2026/04/01/766191/pezeshkian-letter-American-people
6•thatwasunusual•10m ago•0 comments

Graph Based code search that reduces context by 50% in Claude Code

https://github.com/kunal12203/Codex-CLI-Compact
1•TheBengaluruGuy•10m ago•1 comments

Hegseth attempted to make large investments in defense companies before Iran War

https://www.disclose.tv/id/vbilcnkvyw/
3•doener•10m ago•0 comments

Tesla confirms Model S and Model X production is over – only ~600 left

https://electrek.co/2026/04/01/tesla-model-s-x-production-over-only-inventory-left/
2•breve•11m ago•0 comments

Agent Frameworks

https://www.tonykipkemboi.com/blog/agent-frameworks-getting-squeezed
2•tonykipkemboi•12m ago•0 comments

Review the Media You Consume

https://blog.silentvo.id/blog/review-the-media-you-consume/
1•chilipepperhott•12m ago•0 comments

FreeMediaHeckYeah AI

https://fmhy.net/ai
1•Cider9986•12m ago•0 comments

Autofunken

https://airtight.cc/demos/autofunken/
1•memalign•14m ago•0 comments

AWS IAM Language Server

https://github.com/mbarneyjr/aws-iam-language-server
1•mbarneyme•15m ago•1 comments

The Axios NPM compromise and the missing trust layer for AI coding agents

https://digitalegoai.substack.com/p/your-ai-coding-agent-just-installed
1•digitalegoai•16m ago•1 comments

Amazon waives month's AWS charges after Iranian drone attack

https://www.networkworld.com/article/4151880/amazon-waives-entire-months-aws-charges-after-irania...
2•mohit17mor•19m ago•0 comments

Show HN: MurmurCast – Daily briefs from your podcasts, YouTube, and newsletters

https://murmurcast.com
1•Forsheeezy•19m ago•0 comments

Apple at 50

https://www.apple.com/
4•janandonly•20m ago•0 comments

Oracle's 30k-Person Layoff Is a Preview of What's to Come

https://www.hardresetmedia.com/p/the-coldest-cold-email-oracles-30000
3•speckx•20m ago•0 comments

Electrical Transformer Manufacturing Is Throttling the Electrified Future

https://www.bloomberg.com/features/2025-bottlenecks-transformers/
1•toomuchtodo•21m ago•1 comments

Ask HN: What would you like to see in an AI agent?

1•grimm8080•21m ago•0 comments

Anthropic: L'IA che si finge umana

https://www.ircwebnet.com/anthropic-lai-che-si-finge-umana/
1•aktarus•24m ago•0 comments

Codex Plugin for Claude Code

https://github.com/openai/codex-plugin-cc
1•hmokiguess•27m ago•1 comments

For and Against Degrowth

https://fi-le.net/degrowth/
1•fi-le•29m ago•2 comments

Show HN: I Made a Tool to Cleanup Gmail

https://chromewebstore.google.com/detail/apparent-for-gmail/fegjnknjcbfhfgciopbphbojbkmahjeo
2•JaredCampbell•30m ago•1 comments

Comparing C/C++ unity build with regular build on a large codebase (2024)

https://hereket.com/posts/cpp-unity-compile-inkscape/
1•PaulHoule•31m ago•1 comments
Open in hackernews

AI companies charge you 60% more based on your language, BPE tokens

https://tokenstree.com/newsletter-article-5.html
22•vfalbor•1h ago

Comments

vfalbor•1h ago
The Biggest Con of the 21st Century: Tokens How AI Companies Are Charging You More Without You Even Realizing It

You pay for what you use. That's the deal. Except it's not.

When you use an AI model — GPT-4, Claude, Gemini — you do not pay per word. You pay per token. And that tiny technical detail is quietly costing you, depending on which company you choose, up to 60% more for the exact same request.

prophesi•1h ago
Wait until you hear that most models tend to perform worse for non-English languages.
aprentic•44m ago
Do you know if that's true of non-English models?

As I said elsewhere, Deepseek injects Chinese characters into responses. Anecdotally, that seems to happen when the context gets longer. That suggests that they're primarily trained in Chinese and I would expect them to use fewer tokens for Chinese than English.

Mindless2112•1h ago
Funny they didn't include any CJK languages on their list.
bobbiechen•1h ago
I heard an anecdote that Qwen Coder works better when prompted in Korean - haven't tested it for myself though.
aprentic•46m ago
Deepseek will regularly spit out Chinese (汉字)during English sessions. They generally seem to be syntactically related but it makes me think that there's some overhead of using English with an engine that's primarily trained in Chinese.
simianwords•1h ago
This has to be one of the worst things I have read. If this is not satire idk what counts
lxgr•1h ago
“Pay by token” is priced by token, not word or semantic unit; news at 11?

The product itself seems genuinely useful, but the article reads very sensationalist about something that should be pretty obvious.

In other news: French publishers are paying 30% more for paper than English publishers!!

charcircuit•1h ago
The companies didn't arbitrarily choose to bill by tokens. The cost to serve the models scales linearly with tokens which makes it a reasonable pricing strategy. The reality is that you are charged more because it was more expensive to handle the request.
lxgr•1h ago
I guess token length is indirectly determined by language frequency in the training set, and it would be possible to train a model on machine translated training data only to combat that (or maybe to force tokenization to overrepresent languages other than English?), but there’s no way that would be economical, and inference would just be accordingly more expensive to recoup that effort.
Animats•1h ago
It's an ad. "The Solution: TokensTree". From tokenstree.com

I was expecting a secondary market in tokens, perhaps crypto-powered, but no.

The cost difference for languages roughly correlates with how much text it takes to say something in that language. English is relatively terse. (This is a common annoyance when internationalizing dialog boxes. If sized for English, boxes need to be expanded.) They don't list any of the ideographic languages, which would be interesting.

lxgr•1h ago
That would cause the opposite effect of what we’re actually seeing (i.e. “more redundant languages” would be using comparatively fewer tokens).

The real reason is that tokens are probably strictly based on n-gram frequency of the training data, and English is the most common language in the training data.

telotortium•1h ago
My impression of dialog box size from least to greatest is CJK (Chinese < Korean < Japanese) < English < everything else
vfalbor•49m ago
This is not cryto or something else, it’s a platform for tokens reduction. You can try and then post it before do it assumtions. :)
simianwords•1h ago
Europeans be like:

AI commits a racism.

AI commits an environmentalism.

Now use my product (that won't solve either)

simonw•1h ago
The title of this piece differs from the HN title, but the HN title is a lot better. The original title is "The Biggest Con of the 21st Century: Tokens", subhead "How AI Companies Are Charging You More Without You Even Realizing It" - which is an absurd title because tokens are NOT the "biggest con" of anything, and AI companies make it very clear exactly how their pricing works.

I also don't like how this article presents numbers for language differences - in the "The Language Tax" section - but fails to clarify which tokenizer and where those numbers came from.

aprentic•48m ago
There's certainly an interesting question here, even if Tokenstree doesn't provide a solution or even define the problem well.

The broader questions are still interesting.

If an AI is trained more on language A than language B but has some training in translating B to A, what is the overhead of that translation?

If the abilities are combined in the same model, how much lower is the overhead than doing it as separate operations?

ie is f(a) < f(b) < f(t(B,A) ? where a and b are in A and B and f() and t() are the costs of processing a prompt and the cost of translating a prompt.

Then there's the additional question of what happens with character based languages. It's not obvious how it would make sense to assign multiple tokens to a single character but there's the question of how much information in character based vs phonic based words and what the information content of sentences with either one is.