Paradox

Very large part of why I moved away to kagi. It just works.

Additionally, it’s system of weighting, instead of just a binary block, is very useful. Take fandom wikis for example. They’re awful, yes, but sometimes they’re the only result for a topic, and will do if needed. With a binary block list, you either see them or you don’t. With the weighted system, you can downrank them, so if better results show up, they appear higher in the listing than the downranked ones

Can I download their model and run it on my own hardware? No? Then they’re inferior to deepseek