langchainrust 0.19.0

//! Streaming output example
//!
//! Shows how to emit a response token-by-token with `stream_chat`,
//! suitable for real-time display in a chat UI.
//!
//! # Run
//! ```bash
//! cargo run --example basic_streaming
//! ```
//!
//! # Environment variables
//! - `OPENAI_API_KEY`: OpenAI API key (required)
//! - `OPENAI_BASE_URL`: API base URL (optional)

use futures_util::StreamExt;
use langchainrust::schema::Message;
use langchainrust::{BaseChatModel, OpenAIChat, OpenAIConfig};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let api_key = std::env::var("OPENAI_API_KEY")
        .expect("please set the OPENAI_API_KEY environment variable");
    let base_url = std::env::var("OPENAI_BASE_URL")
        .unwrap_or_else(|_| "https://api.openai.com/v1".to_string());

    let llm = OpenAIChat::new(OpenAIConfig {
        api_key,
        base_url,
        model: "gpt-4o-mini".to_string(),
        streaming: true,
        ..Default::default()
    });

    let messages = vec![
        Message::system("You are a helpful assistant."),
        Message::human("Count from 1 to 5."),
    ];

    let mut stream = llm.stream_chat(messages, None).await?;
    let mut total_usage = None;
    while let Some(chunk) = stream.next().await {
        if let Ok(chunk) = chunk {
            print!("{}", chunk.text);
            if chunk.token_usage.is_some() {
                total_usage = chunk.token_usage;
            }
        }
    }
    if let Some(usage) = total_usage {
        println!();
        println!(
            "[token usage] prompt: {}, completion: {}, total: {}",
            usage.prompt_tokens, usage.completion_tokens, usage.total_tokens
        );
    } else {
        println!();
    }

    Ok(())
}