Skip to content

v0.1.0 - Initial Release: ActiveSupport-style LLM Utilities for Ruby

Choose a tag to compare

@patrols patrols released this 17 Feb 08:26
· 2 commits to main since this release
Immutable release. Only release title and notes can be modified.
6100115

🚀 ruby_llm-text v0.1.0 - Initial Release

ActiveSupport-style LLM utilities that make AI operations feel like native Ruby.

✨ Core Features

Transform your Ruby code with intuitive one-liner methods for common LLM tasks:

📝 Text Operations

  • summarize - Condense text with configurable length options
  • translate - Translate between languages with natural language specs
  • extract - Extract structured data using simple schemas
  • classify - Classify text into custom categories

🔧 Developer Experience

  • ActiveSupport-inspired API - Feels like native Ruby methods
  • String extensions - Optional monkey-patching for Rails-like chaining
  • Multi-provider support - Works with OpenAI, Anthropic, Gemini, Ollama via ruby_llm
  • Flexible configuration - Per-method model overrides and settings

📚 Quick Example

require 'ruby_llm/text'

# Configure once
RubyLLM.configure do |config|
  config.openai_api_key = ENV["OPENAI_API_KEY"]
end

# Use everywhere
"Long article text...".summarize(length: :short)
"Bonjour le monde".translate(to: "en")
"John is 30".extract(schema: { name: :string, age: :integer })
"Great product!".classify(categories: ["positive", "negative"])

🏗 Built for the Ruby LLM Ecosystem

Integrates seamlessly with https://github.com/crmne/ruby_llm - inheriting its multi-provider support, configuration system, and robust error handling.


Install: gem install ruby_llm-text
Docs: https://github.com/patrols/ruby_llm-text#readme
Changelog: https://github.com/patrols/ruby_llm-text/blob/main/CHANGELOG.md