A collection of classes to make Ollama API server calls for inference and model management

Getting started

This api client requires a local or remote machine to have Ollama installed and the API server running. It was developed with Flutter 3.13.2 and Dart 3.1.0 on macOS 14.0.

Currently used in one of my other personal projects amallo

Usage

/// 1. Import the package
import 'package:ollama_api_client/ollama_api_client.dart';

/// 2. Create the Ollama Client
final client = OllamaClient();

/// 3. Create a parameter object that describes the request
var parameters = GenerateRequestParams(
    model: 'codellama:34b',
    prompt: 'What are some great features of the dart programming language?',
);

/// 4. Send the request to an Ollama API server
OllamaApiResult result = await client.generateStreamed(parameters);

/// 5. concatenate the response text
if (result.success) {
    String text = '';
    result.data?.listen((GenerateResponse response) {
        text = "$text${response.response}";
        if (response.done) {
            print(text);
        }
      });
}