GptGUI

Python GUI to access Gpt Engine with OpenAI v1.3.3

Handles current specs for:

see https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo
for modle parameters

make sure to pip install the latest openai module
NOTE: Uses OpenAI module >=v1.0.0


Instalation

For Linux / Mac you can either gh repo clone MLeidel/GptGUI or download a zip file of this repo.

Before using this application Python 3.x must be installed.

To install Python go to https://www.python.org/downloads/.
Click Download Python >=3.7.1 and follow the installation instructions.

Use the requirements.txt file to install any modules you may be missing.

    pip3 install -r requirements.txt

You will also have to Sign Up at https://openai.com/api/ and create an API Key.
There is no cost to do so.


(For Text to Speech feature) please install:


After starting the app the first time click the Options button.

Copy and past your API key into the “Gpt Key” entry box.
A better way to handle your API key is to set it up as a System Environment Variable.
Then put the variable’s name into the “Gpt Key” entry box.

Change any other appearance or Gpt options as well.
Then click the Save & Close button.

gptgui.ini

The options are stored in a plain text file called gptgui.ini.
If you prefer you could change the settings with a text editor.

input box


Buttons

alttext

Input your query in the top box, and hit “Submit Query” or Ctrl-g.
The Gpt AI response will appear in the larger box below.

The buttons:


Hot Keys

key action
Ctrl-t View response metrics
Ctrl-m Toggle show-elapsed-time temporarily
Ctrl-h This HotKey help
Ctrl-q Close Program No Prompt
Ctrl-s Save output (Button)
Ctrl-g Submit Query (Button)
Ctrl-Enter Submit Query & Append to existing text
Ctrl-Shift-S Speak the Query Response Text
Escape Cancel Speaking the Text

Context menu

There is a context (Right-Click) menu for convenience. The height of the query text area (top frame) can be lengthened or shortened by in this menu.

How to use the code-davinci-002 Codex model

Read about this here: https://platform.openai.com/docs/models/codex

For GptGUI the top frame acts as the instructions frame and the bottom frame acts as the input frame.


END