BaroqueInMind

I disagree, it made the enemies become tedious damage sponges and currently making the game less enjoyable.

This isn't a Souls game. Whatever difficulty setting is one/two below the hardest is an acceptable balance between completely wasting my time or challenging fun.

Yes. Without a question of a doubt.

I'm self hosting Vaultwarden and my home server got killed by the hurricane, yet I can still access my passwords just fine on the app because it stores them locally encrypted on my phone from the last time it synced. I just can't update or change anything until I can bring everything back on.

So, host your own shit you cowards, it'll be fine.

We don't know if you may be too stupid to have good reading comprehension, but here you go.

Show us where in there it says any North American country? We'll wait.

Yeah, either secure copy (SCP) or rsync are the ways to do this securely.

🤯

That's great, then show me the design plan to construct an F35. I'll wait.

There's no way on Satan's green earth that a small startup company started in 2018 has a search engine index as large and fast as the one started 30 years ago and has been indexing the entire internet since then (Google). Even Microsoft can't compete without dumping billions of dollars into Bing by forcing people to use it as default on their operating systems, renting/purchasing the now ancient Yahoo indexing databases, leeching software engineer talent from competitors with promises of huge paychecks, and greasing hands in China.

In case anyone reading this is curious, Kagi is not capable of storing the impossibly vast and expensive data storage and speeds required to query large indexes of the internet, and so leverages Google and Bing indexes in its search, which is exactly what SearXNG and DuckDuckGo both do for free.

What you pay for is honestly a fancy UI and the ability to quality control the search results for them, again already done for free by other search engines.

Why not use this and select whatever LLM to leverage as a RAG? It literally allows you to self host the model and select any model for both chat and RAG analysis. I have it set to Hermes3 8B for chat and a 1.3B Llama3 as the RAG.