openai-rust

Idiomatic Rust client for the OpenAI API — 1:1 parity with the official Python SDK.

Features

• Async-first (tokio + reqwest)
• Strongly typed requests and responses (serde)
• SSE streaming support
• Automatic retries with exponential backoff
• Builder pattern for requests
• Same resource structure as Python SDK: client.chat().completions().create()

Quick Start

Add to Cargo.toml:

[dependencies]
openai-rust = { git = "https://github.com/fortunto2/openai-rust" }
tokio = { version = "1", features = ["full"] }

use openai_rust::{OpenAI, types::chat::*};

#[tokio::main]
async fn main() -> Result<(), openai_rust::OpenAIError> {
    let client = OpenAI::from_env()?;

    let request = ChatCompletionRequest::new(
        "gpt-4o-mini",
        vec![
            ChatCompletionMessageParam::System {
                content: "You are a helpful assistant.".into(),
                name: None,
            },
            ChatCompletionMessageParam::User {
                content: UserContent::Text("Hello!".into()),
                name: None,
            },
        ],
    );

    let response = client.chat().completions().create(request).await?;
    println!("{}", response.choices[0].message.content.as_deref().unwrap_or(""));
    Ok(())
}

Streaming

use futures_util::StreamExt;
use openai_rust::{OpenAI, types::chat::*};

#[tokio::main]
async fn main() -> Result<(), openai_rust::OpenAIError> {
    let client = OpenAI::from_env()?;

    let request = ChatCompletionRequest::new(
        "gpt-4o-mini",
        vec![ChatCompletionMessageParam::User {
            content: UserContent::Text("Tell me a joke".into()),
            name: None,
        }],
    );

    let mut stream = client.chat().completions().create_stream(request).await?;
    while let Some(chunk) = stream.next().await {
        let chunk = chunk?;
        if let Some(delta) = chunk.choices.first().and_then(|c| c.delta.content.as_deref()) {
            print!("{delta}");
        }
    }
    Ok(())
}

Configuration

use openai_rust::{OpenAI, ClientConfig};

// From environment variable OPENAI_API_KEY
let client = OpenAI::from_env()?;

// Explicit API key
let client = OpenAI::new("sk-...");

// Full configuration
let config = ClientConfig::new()
    .base_url("https://api.openai.com/v1")
    .timeout(30)
    .max_retries(3);
let client = OpenAI::with_config("sk-...", config);

Implemented APIs

More endpoints coming soon: Embeddings, Images, Audio, Files, Models, Fine-tuning, Moderations, Responses.

Development

cargo test                          # all tests
cargo test --features live-tests    # tests hitting real API (needs OPENAI_API_KEY)
cargo clippy -- -D warnings         # lint
cargo fmt -- --check                # format check

License

MIT