https://github.com/jiadingfang/cli-gpt

Simple conversational command line GPT with code interpreter capability that you can run locally with OpenAI API to avoid web usage constraints
https://github.com/jiadingfang/cli-gpt

code-interpreter gpt large-language-models

Last synced: about 2 months ago
JSON representation

Simple conversational command line GPT with code interpreter capability that you can run locally with OpenAI API to avoid web usage constraints

• Host: GitHub
• URL: https://github.com/jiadingfang/cli-gpt
• Owner: jiadingfang
• License: mit
• Created: 2023-05-03T22:42:10.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-11-04T05:39:19.000Z (about 1 year ago)
• Last Synced: 2024-04-14T03:59:09.209Z (9 months ago)
• Topics: code-interpreter, gpt, large-language-models
• Language: Python
• Homepage:
• Size: 20.5 KB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Command Line GPT with Interactive Code Interpreter.

Simple conversational command line GPT that you can run locally with OpenAI API to avoid web usage constraints.

## Update 08/07/23

Add **interactive** code interpreter similar to the web version GPT-4 with Code Interpreter from OpenAI that goes back-and-forth between code generation and GPT reasoning. This is different from [LangChain Python Agent](https://python.langchain.com/docs/integrations/toolkits/python) or [codeinterpreter-api](https://github.com/shroominic/codeinterpreter-api.git) that only generate code one shot. Only focus on python interpreter with numeric usages for now. There are some examples in the `examples` folder.

## Usage

Export `OPENAI_API_KEY` and run `python gpt_dialogue.py` or `python code_interpreter.py`. There are special commands like "save", "pretext", "reset" in the conversation that you can use for convenience.