But how good are these models compared to gpt4o? My last experience with llama2-8b was not great at all. Are there really that good models that would fit on an average consumer hardware (mine has already 32GB ram and 16GB vram)?

The post you're replying to couldn't have made it any easier to answer these questions yourself. No, it won't be as good as the state of the art with massive cloud infrastructure behind an http api.

ChatGPT zealots are still hating on that piece.

Transformers are the taking the cake in the Deep Learning community

Interesting attack vector. Malicious model codes.

Awesome work. It's sad to see the internet being ridden by cancerous ads more and more.

cancerous ads are an integral part of the internet since the dot-com boom

so is ad-blocking

But that's the reality. Doing good on an interview does not guarantee the ability to do the job

"But, but, it's illegal". But nonetheless it was a illuminating debate.

What are the alternarives though?

To Nginx? HAProxy. Or Caddy if you're just doing web stuff.

HAProxy does not serve static files, so it's a poor alternative for this specific case.

I wonder how many concepts that are so complicated for our brain to formalize or even process.