RubyLLM: One Ruby Framework for All Major AI Providers

RubyLLM 1.16.0 is a Ruby framework that provides a unified interface for all major AI providers, including OpenAI, xAI, Anthropic, Gemini, VertexAI, Bedrock, DeepSeek, Mistral, Ollama, OpenRouter, Perplexity, GPUStack, and any OpenAI-compatible API. It supports chat, vision, audio, documents, image generation, embeddings, moderation, tools, agents, structured output, streaming, and Rails integration.

Quick Start

Add to Gemfile: gem 'ruby_llm', then bundle install.

Configure API keys:

RubyLLM.configure do |config|
  config.openai_api_key = ENV['OPENAI_API_KEY']
end

Key Features

• Chat: RubyLLM.chat.ask('question')
• Vision: Analyze images and videos — chat.ask('Describe', with: 'photo.jpg')
• Audio: Transcribe and understand speech — RubyLLM.transcribe('meeting.wav')
• Documents: Extract from PDFs, CSVs, JSON, any file type
• Image generation: RubyLLM.paint('prompt')
• Embeddings: RubyLLM.embed('text')
• Moderation: RubyLLM.moderate('text')
• Tools: Let AI call your Ruby methods by subclassing RubyLLM::Tool
• Agents: Reusable assistants with RubyLLM::Agent
• Structured output: JSON schemas with RubyLLM::Schema
• Streaming: Real-time responses with blocks
• Rails: ActiveRecord integration with acts_as_chat and optional Chat UI
• Async: Fiber-based concurrency
• Model registry: 800+ models with capability detection and pricing

Code Examples

Chat with vision:

chat = RubyLLM.chat
chat.ask('What is in this image?', with: 'ruby_conf.jpg')
chat.ask('Summarize this document', with: 'contract.pdf')

Using tools:

class Weather < RubyLLM::Tool
  desc 'Get current weather'
  def execute(latitude:, longitude:)
    JSON.parse(Faraday.get('https://api.open-meteo.com/...').body)
  end
end
chat.with_tool(Weather).ask('What is the weather in Berlin?')

Structured output:

class ProductSchema < RubyLLM::Schema
  string :name
  number :price
  array :features do
    string
  end
end
response = chat.with_schema(ProductSchema).ask('Analyze', with: 'product.txt')

Rails Integration

bin/rails generate ruby_llm:install
bin/rails db:migrate
bin/rails generate ruby_llm:chat_ui

Then create a Chat model with acts_as_chat and visit /chats for a ready-to-use UI.

RubyLLM has only three dependencies: Faraday, Zeitwerk, and Marcel.