HN user

UrineSqueegee

66 karma
Posts0
Comments84
View on HN
No posts found.
iPhone Air 11 months ago

i was on the fence getting this but I am definitely not after watching this. Probably s26u when it comes out in Jan

GPT-5 12 months ago

pretty underwhelming results so far for me

Open models by OpenAI 12 months ago

I'd really wait for additional neutral benchmarks, I asked the 20b model on low reasoning effort which number is larger 9.9 or 9.11 and it got it wrong.

Qwen-0.6b gets it right.

i opened the ofcom link and it has this really easy to follow guide with 17 illegal contents the users my encounter on your website like terrorism/pdf content etc like extremely bad stuff and all you have to do is asses how likely the user is to run into one of these on your site if its over 0% how do you plan on mitigating that.

thats literally all there is to it.

bro is making 0 sense and I think he fucked something up and he is trying to pin it on the legislation than himself.

it's an extremely trivial thing to do and the ofcom guidance is very easy to follow.

do i have to run each task manually in the tasks.md? can i not leave it unsupervised?

i also need to individually approve each command for some reason and then if it fails due to a service high load i need to manually restart all over again the same task.

i didn't watch the livestream but some people in this thread said that heavy is an orchestration of grok-4s, would be interesting to see how that works