Crate turbine_llm 

Turbine LLM

A unified Rust interface for multiple LLM providers with growing model support.

Turbine provides a simple, consistent API to interact with various Large Language Model providers including OpenAI, Anthropic, Google Gemini, and Groq. Switch between providers with minimal code changes.

Features

• Unified API: Single interface for multiple LLM providers
• Simple & Clean: Minimal, straightforward code
• Text & JSON Output: Support for both text and structured JSON responses
• Async/Await: Built with Tokio for async operations
• Type-Safe: Full Rust type safety with proper error handling

Quick Start

Simplified API (Recommended)

use turbine_llm::TurbineClient;

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create client with model string - automatically detects provider
    let client = TurbineClient::from_model("openai/gpt-4o-mini")?;

    // Simple one-liner to send a message
    let response = client.send("What is Rust?").await?;
    println!("{}", response.content);

    Ok(())
}

Traditional API

use turbine_llm::{TurbineClient, LLMRequest, Message, Provider};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create a client for your chosen provider
    let client = TurbineClient::new(Provider::OpenAI)?;

    // Build a request
    let request = LLMRequest::new("gpt-4o-mini")
        .with_system_prompt("You are a helpful assistant.")
        .with_message(Message::user("What is Rust?"))
        .with_max_tokens(100);

    // Send the request
    let response = client.send_request(&request).await?;
    println!("{}", response.content);

    Ok(())
}

Supported Providers

Set the appropriate API key as an environment variable:

• OpenAI: OPENAI_API_KEY
• Anthropic: ANTHROPIC_API_KEY
• Gemini: GEMINI_API_KEY
• Groq: GROQ_API_KEY

JSON Output

Request structured JSON responses from any provider:

use turbine_llm::{TurbineClient, LLMRequest, Message, OutputFormat, Provider};

let client = TurbineClient::new(Provider::Anthropic)?;

let request = LLMRequest::new("claude-3-5-sonnet-20241022")
    .with_system_prompt("Return data as JSON.")
    .with_message(Message::user("Give me info about Paris"))
    .with_output_format(OutputFormat::Json);

let response = client.send_request(&request).await?;

Re-exports

pub use client::TurbineClient;

pub use error::Result;

pub use error::TurbineError;

pub use models::LLMRequest;

pub use models::LLMResponse;

pub use models::Message;

pub use models::Usage;

pub use types::OutputFormat;

pub use types::Provider;

Modules

client

error

models

providers

types