Generative AI support on Vertex AI is now generally available • Hacker News Board

Re: Generative AI support on Vertex AI is now generally available

#31

Post by itsuka » Sun, Jun 11, 2023, 6:43 AM UTC

zetalabs wrote: post #29

itsuka wrote: post #26

Earlier quoted context omitted.

textembedding-gecko reached GA on June 7th, together with text-bison. You can view all the models available to you on Model Garden.

It doesn't seem to be available to me. How do I get access? If I remember correctly gecko is small enough to run on device.

Are you referring to the text-gecko model? I don't think it's released to the public yet.

> If I remember correctly gecko is small enough to run on device.

Yes, I also remember reading that it's designed to be lightweight enough to run on a mobile device.

Re: Generative AI support on Vertex AI is now generally available

#32

Post by stainablesteel » Sun, Jun 11, 2023, 2:24 PM UTC

Oras wrote: post #4

stainablesteel wrote: post #2

i didn't try it as this requires you to give payment information for a free trial and i got sidetracked what i did learn, is that somehow, google has all of my credit cards despite me never sharing it on the account i was using.

How did you learn that Google has all your credit cards?

it brought me to the payment page for a free trial

Re: Generative AI support on Vertex AI is now generally available

#33

Post by jelling » Sun, Jun 11, 2023, 11:25 PM UTC

twelfthnight wrote: post #22

I've been demoing it and have found it struggles to reliably output structured JSON at the moment. I'm curious if folks have had different experiences and if so what their prompts were.

Guidance from Microsoft or Jsonformer might help. Haven’t tried with vertex but it’s a common problem across LLMs that both projects fix.

Re: Generative AI support on Vertex AI is now generally available

#34

Post by reaperman » Mon, Jun 12, 2023, 7:41 PM UTC

flangola7 wrote: post #18

version_five wrote: post #12

Earlier quoted context omitted.

Presumably they're harder to censor or enforce ideological constraints on. I can't see any other reason other than them being worried about bad press because someone made the model do something that they want to play up as bad.

I can think of two very important reasons just off the top of my head. 1. --- It will kill captchas for good. Half of the internet is protected by Cloudflare or Google captchas at this point. Spam, fraud, and other trouble has a maximum possible volume because you can only pay a human in India so little to solve them for you. If you have an algorithm that can complete it, the game is up. Sites may as well not have a…

> If you have an algorithm that can complete it, the game is up.

This is very much already a thing, I'm sad to say.