Kekw we'll see

Yeah and those are the ones currently identified (btw that issue isn't completely fixed) because rust never was nor advertised itself as sound. Meaning, you gotta be careful when writing Rust code too. Not as much as C++, but it's not a magical shield against memory problems like people have been shilling it as.

A) Rust doesn't have a formal specification other than "whatever the fuck our team hallucinated in this compiler version"

B) Doesn't matter the definition if it fucks your day because you're not careful.

Sure sure Heil Rust but be mindful of the fuck you're doing before you get bit ¯\_ (ツ) _/¯

Have you heard about cve-rs?

https://github.com/Speykious/cve-rs

Blazingly fast memory failures with no unsafe blocks in pure Rust.

Edit: also I wish whoever designed the syntax for rust to burn in hell for eternity

Edit 2: Before the Cult of Rust™ sends their assassins to take out my family, I am not hating on Rust (except the syntax) and I'm not a C absolutist, I am just telling you to be aware of the limitations of your tools

How did you derive the number?

A very specific number?

You do realize how data science works, right? The data is cleared by a human team before it is fed to a model, the data is also normalized and processed by some criteria, etc.

Also if you hold some seriously high value data, if I was designing the system, I'd make it flag your system for more advanced visual retrieval (you can let a multimodal LLM use a website like a human user with tool use)

If you just want to stop scrapers use Anubis why tf are you moving your own goalpost?

Also if your website is trash enough that it gets downed by 15,000 requests you should either hire a proper network engineer or fire yours like wtf man I made trash tier Wordpress websites that handled magnitudes more in 2010

EDIT: And stop using PHP in the case of ScummVM. Jesus Christ this isn't 2005

Not making these famous logical errors

For example, how many Rs are in Strawberry? Or shit like that

(Although that one is a bad example because token based models will fundamentally make such mistakes. There is a new technique that lets LLMs process byte level information that fixes it, however)

Small scale models, like Mistral Small or Qwen series, are achieving SOTA performance with lower than 50 billion parameters. QwQ32 could already rival shitGPT with 32 billion parameters, and the new Qwen3 and Gemma (from google) are almost black magic.

Gemma 4B is more comprehensible than GPT4o, the performance race is fucking insane.

ClosedAI is 90% hype. Their models are benchmark princesses, but they need huuuuuuge active parameter sizes to effectively reach their numbers.

Everything said in this post is independently verifiable by taking 5 minutes to search shit up, and yet you couldn't even bother to do that.

You can experiment on your own GPU by running the tests using a variety of models of different generations (LLAMA 2 class 7B, LLAMA 3 class 7B, Gemma, Granite, Qwen, etc...)

Even the lowest end desktop hardware can run atleast 4B models. The only real difficulty is scripting the test system, but the papers are usually helpful with describing their test methodology.

YO; extremely rare D mention

I discovered digitalmars back in the very early days; it's been on my watchlist for over 15+ years now