Skip to content
TechIndexSubmit
Back to directory

omlx_agents

Twelve production AI agent patterns (structured output, RAG, ReAct, tools, memory, routing...) built on open-source LLMs running locally with oMLX

Similar tools

  • Multi-agent financial news aggregator (fetch, analyze, score, format) running on a local open-source LLM via oMLX

    Models
  • 35 production-grade agentic AI architectures (Reflexion, LATS, GraphRAG, MemGPT, Voyager, BrowserAgent, ...) — a Python library and runnable textbook with multi-provider LLM support and a 17-task benchmark leaderboard.

    Agents
  • Prompts, agent configs, and reference data for using open-weight LLMs to generate automated tests (unit, API, UI/E2E, BDD)

    Models
  • Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.

    Prompts
  • Enables organizations to create and deploy customizable AI agents with support for multiple LLM providers, real-time chat, voice, document training, analytics, user management, SSO authentication, and embeddable widgets. Built with React 19, Redux Toolkit, Radix UI, and Tailwind CSS.

    Dev Tools
  • An open-source, copy-paste pattern library for LLM-generated UIs. The implementation layer for agentic UI standards. While rams.ai, ui-skills.com, and Vercel Design Guidelines define the principles, Agent Patterns provides the ready-to-use React components and LLM-optimized schemas that make those standards actionable.

    UI Kits