Alas. They have said they plan to open some of the source and potentially everything, but it’s little progress.
They recently ported to Linux, which I think will give them much more negative feedback here, so hopefully with more pressure they’ll find the correct copy left license and open up their source to build trust.
There are two modes of AI integrations. The first is a standard LLM in a side panel. It’s search and learning directly in the terminal, with the commands I need directly available to run where I need them. What you get is the same as if you used ChatGPT to answer your questions, then copied the part of the answer you needed to your terminal and run it.
There is also AI Command Suggestion, where you’ll start to type a command / search prefixed by # and get commands directly back to run. It’s quite different from auto-complete (there is very good auto-complete and command suggestion as well, I’m just talking about the AI specific features here).
It’s just a convenient placement of AI at your fingertips when working in the terminal.
Warp.dev! It’s the best terminal I’ve used so far, and the best use of AI as well! It’s extremely useful with some AI help for the thousands of small commands you know exist but rarely uses. And it’s very well implemented.
A non-standardized amount of grams of alcohol in a standard drink.
Each country have their own definitions, usually between 8-14g somewhere, and then each country use that to create their own health rules of how many standard units of alcohol can be part of a healthy nutrition guidelines / low-risk consumption guidelines.
But that’s literally what you do, you build a starter factory to help you build a mid tier research factory which can then help you build a proper factory that can support megascale production
Cool, now do it over Ping!
Ah yeah that’s nowhere even close to registering where I live, luckily.
Ok I’m dumb, ELI5?
Just try it? You’ll get a free trial, use a fake email, no payment details or use info needed as far as I can remember.
I believe also https://kagi.com/fastgpt is free to use.
It’s his distinct style that he has kept since before most of us tried anything like ChatGPT etc, so personally I welcome his consistency.
It’s really cool, but the example doesn’t produce any sensible output? If you have created something like this, why wouldn’t you have your demo output something sensible like Fibonacci or 1337 or whatever.
All these examples are not just using stable diffusion though. They are using an LLM to create a generative image prompt for DALL-E / SD, which then gets executed. In none of these examples are we shown the actual prompt.
If you instead instruct the LLM to first show the text prompt, review it and make sure the prompt does not include any elephants, revise it if necessary, then generate the image, you’ll get much better results. Now, ChatGPT is horrible in following instructions like these if you don’t set up the prompt very specifically, but it will still follow more of the instructions internally.
Anyway, the issue in all the examples above does not stem from stable diffusion, but from the LLM generating an ineffective prompt to the stable diffusion algorithm by attempting to include some simple negative word for elephants, which does not work well.
You have also implemented a ban on any contribution from a VPN without announcing it. It has been very annoying to figure out.
Please do announce these kind of changes and limitations you implement publicly unless you absolutely cannot.
They are plentiful and for rent in Oslo; you can join strangers or book one for your party of guests.
It costs around $18 for a single ticket, or $10 if you’re a member. $250 to rent a floating sauna with room for 12, and options for smaller / larger. All sessions for two hours.
No it isn’t. He for example evaluate that Kagi and Marginalia get the same score if you have to read as far down as to the 10th result for Kagi, while Marginalia has no answer. How is that the same score? There is no explanation. There is a lot of text, and then in the end he has made some subjective choices.
Whoosh?
You can pay with bitcoin, which might help keeping the account entirely anonymous for some.
Yeah, you may be fine with the $5 plan, but that’s the lowest tier available.
Afaik they are not really running a profit yet, just expanding, so that’s an eye opener to how expensive it is to run a search business and how much value Google and others estimate they get from your personal information.
For now though their user base seems fairly much leaning towards business users that can defend this expense as part of becoming more effective professionally. Hopefully over time they’ll grow large enough to provide cheaper plans for regular persons while staying privacy focused and ad free.
Yeah there are some good solutions to achieve this, I’m a big fan of the libredirect project, but currently I’m just setting up the redirects I want directly in Kagi so any URLs in the search results are already rewritten to my liking.
Man I get annoyed just by people eating chips next to me. Yours are horror stories.