machi 0.8.1

A Web3-native AI Agent Framework
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36

//! Agent with vision (multimodal input) example using `OpenAI`.
//!
//! Demonstrates passing an image to an agent via `ContentPart`
//! for visual question answering.
//!
//! ```bash
//! export OPENAI_API_KEY=sk-...
//! cargo run --example agent_openai_vision
//! ```

#![allow(clippy::print_stdout)]

use std::sync::Arc;

use machi::prelude::*;

#[tokio::main]
async fn main() -> Result<()> {
    let provider: SharedChatProvider = Arc::new(OpenAI::from_env()?);

    let agent = Agent::new("vision")
        .instructions("You describe images concisely.")
        .model("gpt-4o-mini")
        .provider(provider);

    // Multimodal input: text + image URL.
    let input = vec![
        ContentPart::text("What is in this image?"),
        ContentPart::image_url("https://picsum.photos/id/237/400/300"),
    ];

    let result = agent.run(input, RunConfig::default()).await?;
    println!("{}", result.output);

    Ok(())
}