chatgpt-functions

This crate provides a wrapper around the OpenAI API to use GPT-3.5 and GPT-4 for chatbots. It also provides a way to define functions that can be called from the chatbot.

Disclaimer

This is a work in progress. The API is not stable and will change.

Requirements

Rust 1.70.0 or higher
OpenAI API key. To get one:
- Go to https://openai.com/
- Sign up
- Go to https://platform.openai.com/account/api-keys
- Create a new API key
- Copy the key and store it in a safe place

Usage

Add the following to your Cargo.toml:

toml [dependencies] chatgpt-functions = "0.1.0"

Documentation

The library is divided in two parts:

chat_gpt contains most of the functions that will be needed in a normal usage:
- ChatGPT is the main struct that contains the context of the chatbot
- completions are the functions that interact with the OpenAI API, storing or not in the context as preferred by the programmer.
The rest of the files contain internal structures and functions that give full flexibility while interacting with the model and the context. But they are not needed for a normal usage. They are public so they can be used if needed.

Examples

You can find examples in the examples folder.

Example without functions

```rust use anyhow::{Context, Result}; use dotenv::dotenv;

use chatgptfunctions::chatgpt::ChatGPT;

[tokio::main]

async fn main() -> Result<()> { dotenv().ok(); let key = std::env::var("OPENAIAPIKEY")?;

let mut gpt = ChatGPT::new(key, None, None)?;

println!("Initialised chatbot. Enter your message to start a conversation.");
println!("Using:");
println!("- Model: {}", gpt.chat_context.model);
println!("- Session ID: {}", gpt.session_id);
println!("You can quit by pressing Ctrl+C (linux), or Cmd+C (Mac).");
println!("--------------------------------------");
loop {
    println!("- Enter your message and press Enter:");
    let mut input = String::new();
    std::io::stdin()
        .read_line(&mut input)
        .context("Failed to read your input")?;
    input.pop(); // Remove the trailing newline

    println!("- AI:");
    // println!("Request: {}", chat_context);
    let answer = gpt.completion_with_user_content(input).await?;
    // println!("Full answer: {}", answer.to_string());
    println!("{}", answer.choices[0].message);
    println!("--------------------------------------");
}

} ```

Example with functions

```rust use anyhow::{Context, Result}; use chatgptfunctions::{ chatgpt::ChatGPT, function_specification::{FunctionSpecification}, }; use dotenv::dotenv;

[tokio::main]

async fn main() -> Result<()> { dotenv().ok(); let key = std::env::var("OPENAIAPIKEY")?;

let mut gpt = ChatGPT::new(key, None, None)?;

let json = r#"
    {
        "name": "get_current_weather",
        "description": "Get the current weather in a given location",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {
                    "type": "string",
                    "description": "The city and state, e.g. San Francisco, CA"
                },
                "unit": {
                    "type": "string",
                    "enum": ["celsius", "fahrenheit"]
                }
            },
            "required": ["location"]
        }
    }
    "#;
let function: FunctionSpecification =
    serde_json::from_str(json).expect("Could not parse correctly the function specification");

gpt.push_function(function);

println!("Initialised chatbot. Enter your message to start a conversation.");
println!("Using:");
println!("- Model: {}", gpt.chat_context.model);
println!("- Session ID: {}", gpt.session_id);
println!("You can quit by pressing Ctrl+C (linux), or Cmd+C (Mac).");
println!("--------------------------------------");
loop {
    println!("- Enter your message and press Enter:");
    let mut input = String::new();
    std::io::stdin()
        .read_line(&mut input)
        .context("Failed to read your input")?;
    input.pop(); // Remove the trailing newline

    println!("- AI:");
    // println!("Request: {}", chat_context);
    let answer = gpt.completion_with_user_content(input).await?;
    println!("Full answer: {}", answer.to_string());
    println!("--------------------------------------");
}

} ```

Example in bash of an interaction with GPT

```bash curl https://api.openai.com/v1/chat/completions -H "Content-Type: application/json" -H "Authorization: Bearer $OPENAIAPIKEY" -d '{ "model": "gpt-3.5-turbo-0613", "messages": [{"role": "system", "content": "You are a helpful assistant."}, {"role": "user", "content": "What is the weather like in Madrid, Spain?"}], "functions": [{ "name": "getcurrentweather", "description": "Get the current weather in a given location", "parameters": { "type": "object", "properties": { "location": { "type": "string", "description": "The city and state, e.g. San Francisco, CA" }, "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]} }, "required": ["location"] } }], "function_call": "auto" }'

{ "id": "chatcmpl-7Ut7jsNlTUO9k9L5kBF0uDAyG19pK", "object": "chat.completion", "created": 1687596091, "model": "gpt-3.5-turbo-0613", "choices": [ { "index": 0, "message": { "role": "assistant", "content": null, "functioncall": { "name": "getcurrentweather", "arguments": "{\n \"location\": \"Madrid, Spain\"\n}" } }, "finishreason": "functioncall" } ], "usage": { "prompttokens": 90, "completiontokens": 19, "totaltokens": 109 } } ```

Contributing

Contributions are welcome! Please open an issue or a pull request.