Crate google_ai_rs

Rust client for Google’s Generative AI APIs

Provides a type-safe, ergonomic interface for interacting with Google’s AI services including Gemini. Features comprehensive API coverage with compile-time schema validation and async/await support.

Highlights

• Type-Safe API Interactions: Generated from official Google discovery documents
• Multi-modal Support: Text, images, and structured data in single requests
• Production-Ready: Connection pooling, retries, and comprehensive error handling

Quickstart

use google_ai_rs::{Client, GenerativeModel};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let client = Client::new("API_KEY".into()).await?;
    let model = client.generative_model("gemini-pro");
     
    let response = model.generate_content(
        "Explain quantum physics using pirate metaphors"
    ).await?;
     
    println!("{}", response.text());
    Ok(())
}

Re-exports

pub use auth::Auth;

pub use client::Client;

pub use error::Error;

pub use genai::GenerativeModel;

pub use crate::proto::Schema;

pub use crate::schema::AsSchema;

pub use crate::schema::SchemaType;

pub use proto::CachedContent;

pub use proto::Candidate;

pub use proto::Content;

pub use proto::FunctionCall;

pub use proto::GenerationConfig;

pub use proto::Part;

pub use proto::TaskType;

pub use proto::Tool;

Modules

auth

chat

client

content

embedding

error

genai

proto

schema

Derive Macros

AsSchema

Warning: No Recursive Types!