News for hackers. Stuff that matters.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by rushingcreek on Thursday February 22, 2024 @02:29PM (#39471838)
Phind founder here. You can try the model for free, without a login, by selecting Phind-70B from the homepage: https://phind.com.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by Fervicus on Thursday February 22, 2024 @03:54PM (#39473079)
I don't use LLMs a lot, maybe once a week or so. But I always pick Phind as my first choice because it's not behind a login and I can use it without giving my phone number. Hopefully you'll keep it that way!
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by worldsayshi on Thursday February 22, 2024 @06:11PM (#39474630)
I don't see how they could. They need to finance it at some point?
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by itsTyrion on Thursday February 22, 2024 @07:34PM (#39475352)
they are already financing it, there are 2 paid plans [0]. For THAT, you need an account (but no phone number).

[0] https://www.phind.com/plans

Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by bbor on Thursday February 22, 2024 @08:40PM (#39475837)
I think there’s room in the market to subsidize real users. Phind delivers absurd value, so I think the majority of paying users could account for the tech-averse or privacy-conscious
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by HKH2 on Friday February 23, 2024 @12:00AM (#39477173)
https://labs.perplexity.ai is the same and it loads much faster than Phind.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by bee_rider on Thursday February 22, 2024 @02:52PM (#39472158)
Important and hard-hitting question from me: have you ever considered calling yourself the Phinder or the Phiounder?
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by Zacharias030 on Thursday February 22, 2024 @03:57PM (#39473122)
or the PhiTO / PhiEO
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by fragmede on Thursday February 22, 2024 @05:14PM (#39474058)
Find Phounder
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by bbor on Thursday February 22, 2024 @08:43PM (#39475860)
Phindational models, phintech, Phinterest, phinder… it might be the best startup name of all time. Hell, startup a password manager and call it Phinders’ Keeper.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by xyzzy_plugh on Thursday February 22, 2024 @09:09PM (#39476068)
Pour one out for Phabricator.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by ComputerGuru on Thursday February 22, 2024 @09:07PM (#39476052)
And here I was wondering why this service was called pee-hind!
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by declaredapple on Thursday February 22, 2024 @03:13PM (#39472487)
Any chances of an API?

And are there plans to release any more weights? Perhaps one or two revisions behind your latest ones?

Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by parineum on Thursday February 22, 2024 @03:14PM (#39472493)
Ask phind to make you one that screen scrapes
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by shrubble on Thursday February 22, 2024 @02:38PM (#39471960)
I tried a question about Snobol4 and was impressed with what it said (it couldn't provide an exact example due to paucity of examples). When testing more mainstream languages I have found it very helpful.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by goldemerald on Thursday February 22, 2024 @02:35PM (#39471919)
Very nice. I've been working with GPT4 since it released, and I tried some of my coding tasks from today with Phind-70B. The speed, conciseness, and accuracy are very impressive. Subjectively, the answers it gives just feel better than GPT4, I'm definitely gonna give pro a try this month.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by visarga on Thursday February 22, 2024 @03:09PM (#39472417)
I prefer Phind's web search with LLM to both Google search and GPT-4. I have switched my default search engine, only using Google for finding sites, not for finding information anymore.

GPT-4 might be a better LLM but its search capability is worse, sometimes sends really stupid search keywords that are clearly not good enough.

Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by bbor on Thursday February 22, 2024 @08:46PM (#39475876)
I won’t steal phind’s thunder but kagi is another great modern tool to have, and much more reliable than google for a technical user IMO. Obviously Phind is irreplaceable for complex or chat-based technical questions, but Kagi sees much more use from me daily for syntax stuff, Wikipedia searches, finding and relating papers, etc.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by browningstreet on Thursday February 22, 2024 @02:37PM (#39471944)
Hmm, when I try I see this in the dropdown:

0 Phind-70B uses left

And I've never made any selection there.

Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by rushingcreek on Thursday February 22, 2024 @02:41PM (#39472005)
I'd suggest logging in in that case -- you will still get your free uses. The Phind-70B counter for non-logged in users has carried over from when we offered GPT-4 uses without a login. If you've already consumed those uses, you'll need to log in to use Phind-70B.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by browningstreet on Thursday February 22, 2024 @02:49PM (#39472118)
Thanks.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by carbocation on Thursday February 22, 2024 @06:17PM (#39474695)
It seems unexpected that other people can edit a link to a Phind chat just by getting the URL. It means that if you share a URL with someone, they can change your results: https://www.phind.com/search?cache=k56i132ekpg43zdc7j5z1h1x
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by airgapstopgap on Thursday February 22, 2024 @06:32PM (#39474843)
Since you're here: have you considered moving to other, better generalist base models in the future? Particularly Deepseek or Mixtrals. Natural language foundation is important for reasoning. Codellama is very much a compromise, it has lost some NLP abilities from continued pretraining on code.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by robbomacrae on Thursday February 22, 2024 @07:02PM (#39475105)
Why do none of the graphs show the speed difference? That seems to be your biggest advantage and the subject line...
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by petesergeant on Thursday February 22, 2024 @07:05PM (#39475134)
This is good stuff, congrats. Took a little detour, but GPT-4 does too (https://www.phind.com/agent?cache=clsxw1mru0033l908mojpvb3b)
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by bobbyi on Thursday February 22, 2024 @07:19PM (#39475228)
I'm selecting 70B and it is coming back with "Answer | Phind-34B Model".

I'm not sure if it's really using the 34B model or if the UI is wrong about which one it used

Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by anter on Thursday February 22, 2024 @07:31PM (#39475332)
You have to click on the "Chat" option at the top left corner, then it'll use the 70B model. I got stuck on that too til I figured that out.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by rushingcreek on Thursday February 22, 2024 @07:32PM (#39475334)
Please try logging in in that case, you will still get your 10 free uses.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by acdanger on Thursday February 22, 2024 @08:32PM (#39475782)
Hi, when I try to use the 70B model from the homepage, the response indicates that it's using the 34B model.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by rushingcreek on Thursday February 22, 2024 @08:35PM (#39475804)
Please try logging in in that case. You will get 10 free daily 70B uses.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by justaj on Thursday February 22, 2024 @08:41PM (#39475850)
Are you considering adding more non-US payment methods for Phind Pro?
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by forevernoob on Friday February 23, 2024 @05:09AM (#39478903)
For sure this. I've recently found out that you can only pay using credit card, US bank account or Cash App.
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by coder1001 on Friday February 23, 2024 @07:22AM (#39479695)
API on the horizon?
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by brainless on Friday February 23, 2024 @09:16AM (#39480828)
Hello Michael, lovely to see this, congrats. Do you already have an API? I could not see it on the site. If not, then do you know around when we can expect it? I am building a desktop BI app with hosted and local LLMs (need schema inference and text to SQL). Would be nice to have Phind as an option for users. Thanks
Re:Phind-70B: Closing the code quality gap with GPT-4 Turbo ... by zestyping on Friday February 23, 2024 @07:54PM (#39488087)
I tried asking "What is the size of Phind-70B's context window?" and it couldn't answer the question. Strangely, it immediately found the page with the answer (https://www.phind.com/blog/introducing-phind-70b) but refused to acknowledge that the answer was there. I tried asking several ways. It even quoted the exact answer in the displayed snippet, but still said there was no answer!

Here are a couple screenshots:

https://imgur.com/a/u7iKOyw https://imgur.com/a/aHAto5H

And here's the link to the whole conversation:

https://www.phind.com/search?cache=zlaksmzkm0h5cpx8l95n62tl

Why is this happening? Does it generally have difficulty with reading web pages, or is there something strange about this particular question?