ollama-rest.rs

Asynchronous Rust bindings of Ollama REST API, using reqwest, tokio, serde, and chrono.

Install

cargo add [email protected]

Features

name	status
Completion	Supported ✅
Embedding	Supported ✅
Model creation	Supported ✅
Model deletion	Supported ✅
Model pulling	Supported ✅
Model copying	Supported ✅
Local models	Supported ✅
Running models	Supported ✅
Model pushing	Experimental 🧪
Tools	Experimental 🧪

At a glance

See source of this example.

use std::io::Write;

use ollama_rest::{models::generate::{GenerationRequest, GenerationResponse}, Ollama};
use serde_json::json;

// By default checking Ollama at 127.0.0.1:11434
let ollama = Ollama::default();

let request = serde_json::from_value::<GenerationRequest>(json!({
    "model": "llama3.2:1b",
    "prompt": "Why is the sky blue?",
})).unwrap();

let mut stream = ollama.generate_streamed(&request).await.unwrap();

while let Some(Ok(res)) = stream.next().await {
    if !res.done {
        print!("{}", res.response);
        // Flush stdout for each word to allow realtime output
        std::io::stdout().flush().unwrap();
    }
}

println!();

Or, make your own chatbot interface! See this example (CLI) and this example (REST API).