![GPT-Code logo](https://private-user-images.githubusercontent.com/1309307/239578698-9ad4061d-2e26-4407-9431-109b650fb022.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.ItBb-zKiRB_XZ7ynaAWjUWflerzM3e62kIxm2wIfr5U)
An open source implementation of OpenAI's ChatGPT Code interpreter.
Simply ask the OpenAI model to do something and it will generate & execute the code for you.
Read the blog post to find out more.
Judah Cooper offered to start & curate a Discord community. Join here.
Open a terminal and run:
pip install gpt-code-ui
gptcode
In order to make basic dependencies available it's recommended to run the following pip
install
in the Python environment that is used in the shell where you run gptcode
:
pip install "numpy>=1.24,<1.25" "dateparser>=1.1,<1.2" "pandas>=1.5,<1.6" "geopandas>=0.13,<0.14" "tabulate>=0.9.0<1.0" "PyPDF2>=3.0,<3.1" "pdfminer>=20191125,<20191200" "pdfplumber>=0.9,<0.10" "matplotlib>=3.7,<3.8"
![GPT-Code logo](https://private-user-images.githubusercontent.com/1309307/239578679-c29c504a-a7ed-4ae0-9360-d7224bc3e3d6.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.s5IHlCiZwGRS0ba_GkBnIXmG_G7HJVUU5umBmSK30kA)
- File upload
- File download
- Context awareness (it can refer to your previous messages)
- Generate code
- Run code (Python kernel)
- Model switching (GPT-3.5 and GPT-4)
You can put a .env in the working directory to load the OPENAI_API_KEY
environment variable.
Set the API_PORT
, WEB_PORT
, SNAKEMQ_PORT
variables to override the defaults.
Set OPENAI_BASE_URL
to change the OpenAI API endpoint that's being used (note this environment variable includes the protocol https://...
).
You can use the .env.example
in the repository (make sure you git clone
the repo to get the file first).
For Azure OpenAI Services, there are also other configurable variables like deployment name. See .env.azure-example
for more information.
Note that model selection on the UI is currently not supported for Azure OpenAI Services.
cp .env.example .env
vim .env
gptcode
localagi took the effort of bundling the Python package in a Docker container. Check it out here: gpt-code-ui-docker.
Please do and have a look at the contributions guide! This should be a community initiative. I'll try my best to be responsive.
Thank you for your interest in this project!