nashex / gpt4-playground Goto Github PK

Is it possible to add a feature where you can export Chat history from Playground mode?

Hello again,

I noticed that when I ask the model for a markdown table the output is in text and it doesn't format to a table structure.

Example:

Here is a table in Markdown format with the top 5 countries and their fictional security levels:

"| Rank | Country     | Security Level |
|------|-------------|----------------|
| 1    | Switzerland | Very High      |
| 2    | Denmark     | Very High      |
| 3    | Finland     | Very High      |
| 4    | Sweden      | High           |
| 5    | Norway      | High           |"

the output should be like an actual table (visually)

Hi,

Thanks again for this wonderful tool.
I noticed that you have two files "OpenAI.ts" and "OpenAIProvider.tsx" , I suppose the second is to keep track of conversation, although, the interface is frozen until the response is completely done then the entire message is printed. I was thinking a good way is to print word by word as they come from the stream.
I tried to edit this but can't seem to solve it.

Hello,

Thanks for this great tool, I am actually trying to adapt your code for Azure OpenAI API, and I am using the chat completion models, here is the API Response :

{
 'id': 'chatcmpl-6p9XYPYSTTRi0xEviKjjilqrWU2Ve',
 'object': 'chat.completion',
 'created': 1677649420,
 'model': 'gpt-3.5-turbo',
 'usage': {'prompt_tokens': 56, 'completion_tokens': 31, 'total_tokens': 87},
 'choices': [
   {
    'message': {
      'role': 'assistant',
      'content': 'The 2020 World Series was played in Arlington, Texas at the Globe Life Field, which was the new home stadium for the Texas Rangers.'},
    'finish_reason': 'stop',
    'index': 0
   }
  ]
}

Can you please tell me how to modify the message object you wrote (id, role, content) so that it takes into account the new reponse.

Thanks,

Hello together
Interesting. I entered the OpenAI Key and it shows only GPT 3 and no GPT 4. You can not select GPT 4.
Why ?
Can you help me there ?
I still think gpt4-playground is great because you don't have to log in. Thanks a lot

Greetings Michael

I want it to function like the regular user version of ChatGPT. In the regular version, ChatGPT will read back previous messages but only to a certain limit, not the entire conversation. This is to avoid situations where the previous conversation is too long and ChatGPT no longer has enough tokens to generate a response, leading to a refusal to respond. Additionally, reading back the entire conversation every time a new message is entered is also very token-consuming."

It seems that it does not support function call.

I am a catering planner, and now I want to make a logo for rice roll shop

In the playground, the system prompt doesn't appear to be saved in the message history. When restoring a message from the history, the system prompt is blank.

Option to add openai key in .env or any file by admin so every user does not need to use keys again and again also add login or authentication system so work for a corporate to provide to their employees

i have my key in env, why wont app us eit?

Hello again,

I noticed that when I'm using the playground interface, and I write a long message, the box keeps scrolling up each time I type a new character at the end.

Thanks.

Hi, I am grateful for the tool because I share my account with others. This allows me to use GPT via API, which is extremely helpful. But since yesterday, I could not be able to use it. The error goes: Unexpected end of JSON input.

Thanks for solving the issue.

I want to run this project locally, how do I start it?

20 minutes ago there was a cross with which you could delete an unnecessary message, but now apparently nothing at all. And as I understand it, after each update, my chat history disappears