ollama

Package Overview

Package selection guidance compiled from public package information, community data, and project usage.

The ollama package enables Dart applications to interact with the Ollama API, facilitating integration with local large language models (LLMs). It provides a straightforward interface for sending prompts and receiving responses, ideal for developers building AI-powered features within cross-platform apps. The package supports all major platforms, including Android, iOS, macOS, Windows, and Linux, making it suitable for both mobile and desktop applications. With minimal setup via Flutter pub add, developers can quickly prototype or deploy generative AI functionality using locally hosted models.

Use cases

Local LLM integration
AI chat applications
Text generation in apps
Code completion tools
Research prototyping

Key features

Cross-platform support
Simple API interface
Supports local model hosting
Real-time response handling
Easy Flutter integration

Suitable for

Flutter developers
AI experimenters
Local LLM users
Full-stack app builders
Privacy-focused applications

Considerations

Requires Ollama server running locally
Limited to Ollama-compatible models
Network configuration needed for remote access
No built-in model management
Dependent on Ollama API stability

Original README

Original README snapshot from the linked source repository. Visit the repository for the latest content.

Ollama for Dart

A Dart client for interacting with the Ollama API. This library provides an easy-to-use interface for generating text completions, chat responses, and embeddings using Ollama inference engine.

Features

Generate text completions
Generate chat responses
Generate embeddings
Support for streaming responses
Customizable model parameters

Installation

Run the following command to install the package:

dart pub add ollama

Usage

Initializing the client

import 'package:ollama/ollama.dart';

final ollama = Ollama();
// Or with a custom base URL:
// final ollama = Ollama(baseUrl: Uri.parse('http://your-ollama-server:11434'));

Generating text completions

final stream = ollama.generate(
  'Tell me a joke about programming',
  model: 'llama3',
);

await for (final chunk in stream) {
  print(chunk.response);
}

Generating chat responses

final messages = [
  ChatMessage(role: 'user', content: 'Hello, how are you?'),
];

final stream = ollama.chat(
  messages,
  model: 'llama3',
);

await for (final chunk in stream) {
  print(chunk.message?.content);
}

Generating embeddings

final embeddings = await ollama.embeddings(
  'Here is an article about llamas...',
  model: 'llama3',
);

print(embeddings);

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Search packages, repositories and app categories

Install package