Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

github.com

Post reply on HN
Print view View original

81–90 of 102 posts

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#81

Post by squeaky-clean »

hesdeadjim wrote: post #49
literalAardvark wrote: post #39

Earlier quoted context omitted.

The real problem with 4 is that OpenAI are having trouble keeping up with the demand. So it might not be an option until 4 turbo comes out in 37 hours or whatever their development cycle is these days.

Yea, I find it a fair bit annoying that I pay $20 and have a 25 message cap every 3 hours.

I don't necessarily mind the cap, but I am a little annoyed that it was 50 messages every 4 hours when I first began my subscription and they lowered it without so much as an email telling me.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#82

Post by totoglazer »

arthurcolle wrote: post #79
ukuina wrote: post #52

Earlier quoted context omitted.

I use TypingMind.com and the GPT4 API. Seems an overall better experience.

GPT-4 API still not open. I should really try Evals to get early access, been dying to give it a spin (outside of ChatGPT)

It works for me.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#83

Post by Closi »

squeaky-clean wrote: post #81
hesdeadjim wrote: post #49

Earlier quoted context omitted.

Yea, I find it a fair bit annoying that I pay $20 and have a 25 message cap every 3 hours.

I don't necessarily mind the cap, but I am a little annoyed that it was 50 messages every 4 hours when I first began my subscription and they lowered it without so much as an email telling me.

I guess you have experienced something different, as I got an email saying this:

> GPT4 will have a dynamically adjusted usage cap. We expect to be severely capacity constrained, so the usage cap will depend on demand and system performance. API access will still be through the waitlist.

And then the actual 'sign-up' or marketing page before subscription didn't even say anything about getting GPT-4 (just about getting priority access to the standard chatgpt product).

Then at the bottom of each page above the box when you open GPT-4 as long as I remember it has always said something like "Current limit is this, capacity limits for GPT4 will be changed as we adjust for demand."

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#84

Post by Cyphase »

alach11 wrote: post #77
Cyphase wrote: post #12

Earlier quoted context omitted.

(I may have misunderstood your comment to some extent, but I'm going to send this reply anyway even if just to clarify for anyone else who might misunderstand.) --- I agree with "be careful what you send to the chat bot", but let's clarify some things in case you or someone else reading your comment is misunderstanding. LLMs aren't immature AI brains that "may even permanently ingest user data for training purposes".…

> If you're talking about OpenAI specifically saving ChatGPT data for later training purposes, they absolutely are doing that They claim they're not retaining data through the API.

That's correct (after 30 days). When I say ChatGPT I mean the web-based frontend product, not the models behind it which you can also access via API.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#85

Post by Cyphase »

fl7305 wrote: post #59
Cyphase wrote: post #12

Earlier quoted context omitted.

(I may have misunderstood your comment to some extent, but I'm going to send this reply anyway even if just to clarify for anyone else who might misunderstand.) --- I agree with "be careful what you send to the chat bot", but let's clarify some things in case you or someone else reading your comment is misunderstanding. LLMs aren't immature AI brains that "may even permanently ingest user data for training purposes".…

> Running inference does not change the trained weights. That's true today. I don't know how many days (or hours) away we are from GPT-4 running a LoRA-pass to update its weights after each round though?

That would be an explicit decision by OpenAI, not a result of immature tech.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#86

Post by Cyphase »

semiquaver wrote: post #15
Cyphase wrote: post #12

Earlier quoted context omitted.

(I may have misunderstood your comment to some extent, but I'm going to send this reply anyway even if just to clarify for anyone else who might misunderstand.) --- I agree with "be careful what you send to the chat bot", but let's clarify some things in case you or someone else reading your comment is misunderstanding. LLMs aren't immature AI brains that "may even permanently ingest user data for training purposes".…

from the “reliably isolate sessions” reference I have to assume they are referring to this: https://news.ycombinator.com/item?id=35291112

I figured the same; I didn't address that point.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#87

Post by ericmcer »

literalAardvark wrote: post #39
jbellis wrote: post #29

Earlier quoted context omitted.

GPT4 is so much better than 3.5 at virtually everything that I don't think it's worth trying to figure out which ones 3.5 is almost adequate enough for. Also, the pricing is per token, so even with 4 it is close to negligible unless you are loading in a lot of context or your conversation gets very long.

The real problem with 4 is that OpenAI are having trouble keeping up with the demand. So it might not be an option until 4 turbo comes out in 37 hours or whatever their development cycle is these days.

Seriously their release cycle has been so rapid that I am building stuff with the idea that AI will be better and cheaper by the time I am ready to release it.

Right now it is a bit of a blocker because you can easily get a single prompt to cost you .005c - .01c, which would crush you if you ever had any kind of scale.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#88

Post by LesZedCB »

squeaky-clean wrote: post #81
hesdeadjim wrote: post #49

Earlier quoted context omitted.

Yea, I find it a fair bit annoying that I pay $20 and have a 25 message cap every 3 hours.

I don't necessarily mind the cap, but I am a little annoyed that it was 50 messages every 4 hours when I first began my subscription and they lowered it without so much as an email telling me.

well it was something like 100 before that even.

Re: Personal Concierge Using OpenAI's ChatGPT via Telegram and Voice Messages

#89

Post by TeMPOraL »

totoglazer wrote: post #82
arthurcolle wrote: post #79

Earlier quoted context omitted.

GPT-4 API still not open. I should really try Evals to get early access, been dying to give it a spin (outside of ChatGPT)

It works for me.

Because you were one of the blessed lucky ones to get access to GPT-4 API? I'm still on the waitlist :(.
Post reply on HN