Crate agentkit_provider_ollama 

Ollama model adapter for the agentkit agent loop.

This crate provides OllamaAdapter and OllamaConfig for connecting the agent loop to a local Ollama instance via its OpenAI-compatible chat completions endpoint. It is built on the generic agentkit_adapter_completions crate.

No API key is required — Ollama runs locally and does not authenticate requests by default.

Quick start

use agentkit_loop::{Agent, SessionConfig};
use agentkit_provider_ollama::{OllamaAdapter, OllamaConfig};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Ollama must be running locally (e.g. `ollama serve`).
    let config = OllamaConfig::new("llama3.1:8b");
    let adapter = OllamaAdapter::new(config)?;

    let agent = Agent::builder()
        .model(adapter)
        .build()?;

    let mut driver = agent
        .start(SessionConfig::new("demo"))
        .await?;
    Ok(())
}

Structs

OllamaAdapter

Model adapter that connects the agentkit agent loop to a local Ollama instance.

OllamaConfig

Configuration for connecting to a local Ollama instance.

OllamaProvider

The Ollama provider, implementing CompletionsProvider.

OllamaRequestConfig

Request parameters serialized into the Ollama request body.

Enums

OllamaError

Errors produced by the Ollama adapter.

Type Aliases

OllamaSession

An active session with a local Ollama instance.

OllamaTurn

A completed turn from a local Ollama instance.