zmolecristian / openai-gptokenomics

A command line tool to get the price of the request sent to OpenAI models

OpenAI-GPTokenomics

OpenAI-GPTokenomics is a simple command-line tool to calculate the price for requests sent to the OpenAI models.

Features

• Calculate request price based on model and number of tokens.
• Calculate answer price based on model and number of tokens.
• Support for multiple OpenAI models: GPT-4, GPT-3.5 Turbo, DaVinci, Curie, Babbage, and Ada.
• Read data from input files or directly from command-line arguments.
• Write output to a file or print it to the console.

Requirements

• Rust (tested with Edition 2021)
• Dependencies:
  • tiktoken-rs v0.3.3
  • structopt v0.3

Installation

• Clone the repository:

git clone https://github.com/tragDate/openai-gptokenomics.git

• Change into the project directory:

cd openai-gptokenomics

• Build the project:

cargo build --release

• The binary will be available at ./target/release/openai-gptokenomics.

Usage

openai-gptokenomics [FLAGS] [OPTIONS]

Flags:

• -r, --request_file: A file with the request data (conflicts with --request_data).
• -d, --request_data: A string with the request data (conflicts with --request_file).
• -a, --answer_file: A file with the answer data (conflicts with --answer_data).
• --answer_data: A string with the answer data (conflicts with --answer_file).
• --answer: Calculate answer price from stdin.
• -o, --output_file: Save the output into a file, instead of printing it to the console.

Options:

• -m, --model (default: gpt-3.5-turbo): The OpenAI model.
• --output_file: The path to save the output.

Example:

openai-gptokenomics -m "gpt-4" -d "What is the capital of France?" --answer_data "The capital of France is Paris."

Author

@tragDate

License

GPL-3.0