A very variant I really like is using Flux Schnell. I use it on replicate.com but you can find it for free all over the web such as https://huggingface.co/black-forest-labs/FLUX.1-schnell

Result for "color palette of 8 colors for a china-themed site. Include opposing complementary colors."

https://replicate.com/p/zrb89ym2r1rm80ckxf1995yxcr

From what I've come to understand:

- Romania's voting system is in 2 rounds, and he won the first. The second is in ~2 days.

- Most people never heard of him up until 2 weeks before the first round, when an aggressive Tiktok campaign started. Polymarket.com didn't include him in their system. Many were taken by surprise.

- Many videos promoting him on Tiktok weren't marked as being propaganda, and thus allowed to go viral.

- He claims 0 campaign spending and that it's thanks to volunteers.

- Many who voted for him aren't into politics, and liked the way he presented himself.

- His content is a mix of populist, nationalist, anti-establishment, and mostly contain patriotic and emotional narratives to connect to viewers.

- Larger cities voted for the pro-western candidate

- He's made some controversial statements, such as the moon landing being fake, covid denying (am not sure to what extent), 5G conspiracy, denying climate change (those being just a few)

Everyone's talking about him. I'm curious how it will turn out and to see how the quote "there's no such thing as bad publicity" fares in this case.

That's an excellent summary. I'd add: - Someone so unexpectedly winning the first round creates its own myth, and people tend to join such candidates "to be on the winner's side", so he seemed to have a solid chance at winning the second round too.

Thank you for clarifying. I also initially thought it was an (LLM (vulnerability scanner)).

must be a reflection how people are thinking; since I'm infosec oriented, I interpreted it as ((LLM vulnerability) scanner)

It does edit files directly when using its' Composer feature (Ctrl/Cmd+I).

I agree, I'm also firmly convinced it's the best. Until a month ago it has some bugs I disagreed with some design choices, but since 1-2 weeks ago it's the absolute best I've ever used and have easily convinced others to use it.

Now that you've pointed it out I've also realized there was something off about how I thought about attack helicopters when I first read the title.

Same here. I think normally I'd move on, but because it's a Show HN and it got some attention I'm forcing myself.

They're likely referring to using the large language models (GPT, Claude, etc.) through their primary interfaces, because you get more control by interacting with the LLM directly.

I have been eyeing your course for a while now. This is fantastic! Thank you so much!