🦀 Embeddings
SurrealDB offers comprehensive support for vector embeddings, enabling powerful semantic search and machine learning capabilities across your data. Through integrations with leading embedding providers, you can easily store, index and query high-dimensional vectors alongside your regular data.
Mistral
use mistralai_client::v1::{client::Client, constants::EmbedModel};
static KEY = std::env::var("MISTRAL_API_KEY").unwrap();
let client = Client::new(Some(KEY.to_string()), None, None, None)?;
let input = vec!["Joram is the main character in the Darksword Trilogy.".to_string()];
let result = client.embeddings_async(MODEL, input, None).await?;
println!("{:?}", result);
Find a full example in Semantic search in Rust with SurrealDB and Mistral AI.
Ollama
use ollama_rs::{Ollama, generation::embeddings::GenerateEmbeddingsRequest};
#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
let ollama = Ollama::default();
let model = "all-minilm:22m".to_string()
let prompt = "this is your input text".to_string();
let request = GenerateEmbeddingsRequest::new(model, prompt);
let response = ollama.generate_embeddings(request).await?;
println!("Generated embeddings (first 5): {:?}", &response.embeddings[..5]);
println!("Embedding vector length: {}", response.embeddings.len());
Ok(())
}
use rust_bert::sentence_embeddings::{
SentenceEmbeddingsBuilder, SentenceEmbeddingsModelType,
};
fn main() -> anyhow::Result<()> {
let model = SentenceEmbeddingsBuilder::remote(
SentenceEmbeddingsModelType::AllMiniLmL6V2
).create_model()?;
let sentences = [
"this is your text",
"you can encode more than one in batch"
];
let embeddings = model.encode(&sentences)?;
for (i, embedding) in embeddings.iter().enumerate() {
let truncated_embedding: Vec<_> = embedding.iter().take(5).cloned().collect();
println!("\nSentence: '{}'", sentences[i]);
println!("Embedding (first 5 values): {:?}", truncated_embedding);
println!("Embedding dimensions: {}", embedding.len());
}
Ok(())
}
