prediction-guard 0.5.0

SDK for the Prediction Guard API
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

//! `completion` sends a prompt to Prediction Guard and returns a single reponse of
//! type [`completion::Response`].
extern crate prediction_guard as pg_client;
use pg_client::{client, completion};

#[tokio::main]
async fn main() {
    let pg_env = client::PgEnvironment::from_env().expect("env keys");

    let clt = client::Client::new(pg_env).expect("client value");

    let req = completion::Request {
        model: completion::Models::NeuralChat7B,
        prompt: "Will I lose my hair?".to_string(),
    };

    let result = clt
        .generate_completion(&req)
        .await
        .expect("completion response");

    println!("\ncompletion response:\n\n{:?}", result);
}