jas3333 / gpt-helper Goto Github PK

When i write a prompt it will continue to write as me. For example i said: "Yep" as a response. The Program Took my "Yep" and made it "You: Yep" at the top and the response was. "That's about it". The thing is that it is just continuing what i was saying instead a response from the ai. Also The ai just stop responding to me at all because it has no indication that it is its turn to talk. I think if you add a "response:" In front of what the ai says it will fix the issue of not knowing who is talking.

Hey! awesome work. I've gone and forked it and made it usable on mobile: https://github.com/TrySpace/GPT-Helper

And it uses typescript and @mui components, I've also cleaned up almost all of the unused css, and other files.

PR's welcome for improvements, would love to see it merged here too, although I've changed the folder structure a bit (no need for client folder)

/Edit oops I see that there was a big update in the meantime, so my fork is based on a few weeks older version 😅

Hello! Sorry for bothering you, but maybe you'll be able to help me solve problem that appeared recently.

The app used to work before, now it launches UI and just ignores anything I type, no matter which language or model or settings I use. Another app of yours, GPT-Chat, works just fine. API Key is the same, and it's located in the correct place (according to the ‘install’ section). Also ChatGPT web page works just fine, so it's not like my account blocked or restricted for some reason. Do you know what might be causing this behaviour? Where might I look for error log (to figure out what goes wrong)?

OS: MacOS Ventura 13.2
Laptop: MacBook Air Late 2020 with an M1 processor.

P.S. Solved by creating new API on openAI site 🤷

Just keep getting 'Incorrect API key provided: undefined. You can find your API key at https://platform.openai.com/account/api-keys.'

Have just created an account with openAI and 2 API keys, but both have resulted in the above error.

in the .env file I have tried

REACT_APP_OPENAI_KEY = 'sk-8dL......' & REACT_APP_OPENAI_KEY = sk-8dL......

Any help to get this running is much appreciated.

The floating input field can end up covering up the last response once a page is full and it starts scrolling.

Chrome(Version 110.0.5481.100 (Official Build) (64-bit) ) on Linux Kubuntu.

Hey this is a greate project! I set this up and it runs but I can't seem to ask any questions? I'm using Firefox - am I doing something wrong? When I press enter after asking a question nothing happens...Any help appreciated

hi

can you please make the main GPT-Helper branch like docker version working with Pinecone and Mongodb .

regards

Hello and thank you very much for this easy tool you are providing.

I was wondering if you are thinking to bring it to mobile too in the future.
Or if you know a valid solution for mobile that does the same as your program.

Thanks

The model's maximun content length is 4097 tokens, it means that you can write only 4097 words. Is there any way to increase this number?

After some time chatting the Chatbot randomly stops responding and my typed text disappears.

Will this still operate as normal while the ChatGPT is down/overloaded?

Thanks for updating the code but I kinda liked or was used to the 'npm start' from the client directory which not only started the GPT-Helper server but also spawned a browser tab loaded to the chat server.

The current method of starting does work and I was able to write a bash script to do what the previous client did but it's not as elegant.

#!/bin/bash
HERE=pwd
cd ~/GPT-Helper/
google-chrome http://localhost:4001
node server.js
cd $HERE

hi

can you please add gpt-3.5-turbo new model
its the most capable GPT-3.5 model and optimized for chat at 1/10th the cost of text-davinci-003. its updated with latest model iteration.

thanks in advance

When I asked what today's date was it replied with "Today is Wednesday, April 7th, 2021.".
My system clock is set via ntp and is accurate. Is this getting the date/time from the past? ;-)

I think i've set it up properly. I have the UI running and I also have a GPT.env file on my client folder (does the filename matter? the tutorial said only .env) my env file has my open AI API key attached.

Why is this not working?

Hi using Windows / Anaconda I've managed to install it but what Command do I use to input my API Code?

If I use REACT_APP_OPENAI_KEY=mykey I get an error

" is not recognized as the name of a cmdlet,
function, script file, or operable program. Check the spelling of the name, or if a path was included, verify that the
path is correct and try again."