Struct GeminiClient 

pub struct GeminiClient { /* private fields */ }

HTTP client for interacting with Google’s Gemini Embeddings API.

This client provides methods to generate text embeddings using Google’s gemini-embedding-001 model. Embeddings are vector representations of text that can be used for semantic search, clustering, and similarity comparisons.

Security

The API key is securely transmitted via the x-goog-api-key HTTP header, not in the URL, to prevent accidental exposure in logs and proxies.

Examples

use ceres_client::GeminiClient;

let client = GeminiClient::new("your-api-key")?;
let embedding = client.get_embeddings("Hello, world!").await?;
println!("Embedding dimension: {}", embedding.len()); // 768

Implementations

impl GeminiClient

pub fn new(api_key: &str) -> Result<Self, AppError>

Creates a new Gemini client with the specified API key.

pub async fn get_embeddings(&self, text: &str) -> Result<Vec<f32>, AppError>

Generates text embeddings using Google’s gemini-embedding-001 model.

This method converts input text into a 768-dimensional vector representation that captures semantic meaning.

Arguments

• text - The input text to generate embeddings for

Returns

A vector of 768 floating-point values representing the text embedding.

Errors

Returns AppError::ClientError if the HTTP request fails. Returns AppError::Generic if the API returns an error.

pub async fn get_embeddings_batch( &self, texts: &[&str], ) -> Result<Vec<Vec<f32>>, AppError>

Generates embeddings for multiple texts in a single API call.

Uses the batchEmbedContents endpoint which supports up to 100 texts.

Arguments

• texts - Slice of text references to embed

Returns

A vector of embedding vectors, one per input text, in the same order.

Trait Implementations

impl Clone for GeminiClient

fn clone(&self) -> GeminiClient

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl EmbeddingProvider for GeminiClient

fn name(&self) -> &'static str

Returns the provider identifier for logging and configuration.

fn dimension(&self) -> usize

Returns the embedding dimension this provider generates.

fn max_batch_size(&self) -> usize

Maximum number of texts supported per batch API call.

async fn generate(&self, text: &str) -> Result<Vec<f32>, AppError>

Generates an embedding vector for the given text.

async fn generate_batch( &self, texts: &[String], ) -> Result<Vec<Vec<f32>>, AppError>

Generates embeddings for multiple texts in a batch.