This Ruby gem provides a low-level interface to interact with Ollama's API, enabling developers to run open-source Large Language Models (LLMs) locally. It serves as a foundational tool for building custom AI-powered applications and abstractions on top of Ollama, targeting Ruby developers who need direct control over LLM interactions.

How It Works

The gem acts as a direct client for Ollama's REST API, exposing methods that map to Ollama's core functionalities. It handles request serialization, response parsing, and streaming capabilities via Server-Sent Events (SSE). This approach offers granular control over model interactions, including text generation, chat completions, embeddings, and model management, without introducing higher-level abstractions that might limit flexibility.

Quick Start & Requirements

• Installation:
  • Add to Gemfile: gem 'ollama-ai', '~> 1.3.0'
  • Or install directly: gem install ollama-ai -v 1.3.0
• Prerequisites:
  • A running Ollama instance (defaulting to http://localhost:11434).
  • Ruby environment.
• Links:
  • Ollama Official Website: https://ollama.ai/
  • Ollama GitHub: https://github.com/jmorganca/ollama
  • Ollama API Documentation: https://github.com/jmorganca/ollama/blob/main/docs/api.md

Highlighted Details

• Comprehensive API coverage: Supports text generation, chat, embeddings, and model management (create, tags, show, copy, delete, pull, push).
• Streaming support: Integrates with Server-Sent Events (SSE) for real-time responses.
• Image input: Allows processing images by encoding them as Base64 strings.
• Customizable client: Supports configuring connection adapters, timeouts, and error handling.

Maintenance & Community

The README details development tasks and publishing procedures but does not list specific maintainers, community channels (like Discord/Slack), or a public roadmap.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Compatible with standard Ruby environments. The MIT license permits commercial use, with a disclaimer of warranty and no assumption of responsibility for damages.

Limitations & Caveats

This gem provides low-level access and requires developers to build their own higher-level abstractions. It is dependent on a separately running Ollama service. Image processing requires manual Base64 encoding and may necessitate increased client timeouts.