Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

RubyLLM

Ruby-native AI framework with one API across 19 providers, including a TypeSafe provider for Jev's typed judgments.

LibraryproductionMIT
View on GitHub ↗

Overview

RubyLLM is a Ruby-native AI framework for working with models, tools and agents through one API in plain Ruby or Rails. It supports chat with files (images, video, audio, PDFs, code), streaming, image and video generation, embeddings, reranking, transcription and more across 19 providers, including OpenAI, Anthropic, Gemini, Bedrock, Ollama, OpenRouter and TypeSafe, plus any OpenAI-compatible endpoint.

TypeSafe is listed among the providers. The list summary describes a native System One protocol for typed questions, probabilistic answers and error normalisation, and a related gem's README mentions upcoming RubyLLM judgments (RubyLLM::Judge classes with probability, choice and score questions answered by Jev through a built-in TypeSafe provider). The fetched README excerpt does not show the judgment API itself.

What the decision model decides

TypeQuestionAnswers
OtherTyped probability, choice and score judgments answered by Jev through the TypeSafe provider(inferred)—

How it works

  1. Configure the TypeSafe API key in RubyLLM
  2. Define typed questions (judgments)
  3. Send to Jev through the TypeSafe provider
  4. Use probabilistic answers in Ruby code

Strengths

  • Same framework for generative and decision models
  • Mature, widely used Ruby library

Caveats

  • Jev specifics not shown in the fetched README
  • Judgments may depend on a newer RubyLLM release

At a glance

Type
Library
Status
production
Language
Ruby
Runtime
Ruby gem / Rails
Has code
yes
Has tests
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • Ruby ecosystem: official Ruby gem connecting TypeSafe judgment models to RubyLLM with a native System One protocol for typed questions, probabilistic answers, and error normalization.

    yibie/awesome-jev

Frequently asked questions

Does RubyLLM support Jev?

TypeSafe is one of its listed providers, and list sources describe native System One support.

What are RubyLLM judgments?

Per a related gem's README, Judge classes with probability, choice and score questions answered by Jev.

Is it Rails-friendly?

Yes, it works in plain Ruby or Rails.

Related projects

  • Ax

    Ax, a DSPy-style multi-language LLM framework, listed with a native TypeSafe provider for Boolean, Choice, Score and raw Jev questions.

    PluginproductionTypeScript
  • Effect

    Effect TypeScript library whose @effect/ai-typesafe package maps classify, probability and rating operations to Jev questions.

    ProjectproductionTypeScript
  • instructor-php

    PHP monorepo whose Polyglot module adds a TypeSafe driver for typed Noul, Choice and Score decisions alongside LLM inference.

    ProjectproductionPHP
  • req\_llm

    Elixir library for unified LLM calls across providers, with an evaluate interface that calls Jev via TypeSafe or OpenRouter.

    ProjectproductionElixir
  • synkora-ai

    Self-hosted multi-tenant AI agent platform listed as including optional TypeSafe client tools for classification, scoring and yes/no.

    LibraryproductionPython
  • TanStack AI

    Provider-agnostic TypeScript AI SDK; per list sources, an @tanstack/ai-typesafe adapter exposes Jev decisions through decide().

    ServiceproductionTypeScript