crmne/ruby_llm

The Ruby AI framework.

What it solves

RubyLLM is a native Ruby framework designed to simplify the integration of AI capabilities into Ruby and Rails applications. It provides a unified API to interact with 19 different AI providers, allowing developers to switch between hosted and local models without rewriting their application logic.

How it works

The framework acts as an abstraction layer over various AI providers (such as OpenAI, Anthropic, Gemini, and Ollama). It allows developers to use a single consistent interface for tasks like chatting, image and video generation, audio transcription, and OCR. It supports advanced AI patterns including tool use (function calling), agentic loops, and structured output via Ruby schemas.

Who it’s for

Ruby and Rails developers who want to build AI-powered features—such as conversational agents, document analysis, and multimodal AI—without being locked into a specific provider's SDK.

Highlights

  • Unified Provider API: Support for 19 providers including OpenAI, Azure, xAI, Anthropic, Gemini, VertexAI, Bedrock, Cohere, DeepSeek, and Ollama.
  • Multimodal Capabilities: Built-in support for analyzing images, videos, and audio, as well as generating images and videos.
  • Agentic Framework: Tools for creating reusable agents with specific instructions and tools, including human-in-the-loop approval workflows.
  • Rails Integration: Deep integration with Active Record for persistence, Active Storage for attachments, and Hotwire for streaming responses.
  • Evaluation Suite: A system for testing agent correctness using datasets and model-based evaluators.
  • Developer Tooling: Includes prompt templates (ERB), cost tracking, token counting, and prompt caching.

Related

  • Dispatch
    RubyLLM: A Unified Framework for AI Providers in Ruby

    RubyLLM is a comprehensive Ruby framework that provides a single, unified interface for interacting with major AI providers including OpenAI, Anthropic, Gemini, and local models via Ollama.

  • Project
    activeagents/activeagent

    Active Agent is a Ruby on Rails gem that introduces an agent‑oriented programming model for building AI‑powered features. Agents act like controllers, support multiple LLM providers (OpenAI, Anthropic, Ollama, etc.), offer prompt templating, streaming, tool calling, delegation, and optional telemetry via the separate `actionagent` dashboard gem.

  • Project
    BoxcarsAI/boxcars

    A Ruby framework for building LLM-powered systems that allows developers to coordinate tools, data queries, and AI providers through a unified, composable architecture.

  • Project
    ParisNeo/lollms-webui

    LoLLMs WebUI is an open‑source, locally‑run web interface that lets you chat with, route, and combine dozens of LLM and multimodal models (text, image, video, music). It supports many back‑ends (Hugging Face, Ollama, OpenAI, etc.), offers “personality” agents for specific tasks, and includes a smart routing system that chooses the cheapest or fastest model for each prompt. Install via Python virtualenv or Docker, configure bindings, and launch `app.py` to start the UI.

  • Project
    laravel/ai

    A unified AI SDK for Laravel that provides a consistent API to interact with providers like OpenAI, Anthropic, and Gemini for agents, images, and audio.