Running Mixture of Agents on CPU: LFM2.5 Brain (1.2B) + Falcon-R Reasoner (600M) + Tool Caller (90M). CPU-only, 16GB RAM. Lightweight AI Legion.

Official code of Attention-MoA: Enhancing Mixture-of-Agents via Inter-Agent Semantic Attention and Deep Residual Synthesis

LLMProxy is an intelligent large language model backend routing proxy service.

Tribunal multi-agent Discord bot framework — autonomous paired review with configurable agent roles

An external pipeline filter for Open WebUI that refactors aggregate requests with collective wisdom to output structured expert analysis reports.

This Streamlit application showcases the Mixture of Agents (MOA) architecture proposed by Together AI, powered by Groq LLMs. It allows users to interact with a configurable multi-agent system for enhanced AI-driven conversations.

GPS for your LLMs

A simplified agentic workflow process based on the Mixture of Agents (MoA) system for Large Language Models (LLMs)

A mixture of agents allowing you to combine responses from different AI models.

High-performance On-Device MoA (Mixture of Agents) Engine in C++. Optimized for CPU inference with RadixCache & PagedAttention. (Tiny-MoA Native)

This is the official repository for the paper: This is your Doge: Exploring Deception and Robustness in Mixture-of-LLMs.

Production-ready Phidata AI agent implementations covering 16+ use cases: data analysis, financial advisory, healthcare diagnostics, content generation, web research, and multi-agent systems. Perfect starter kit for developers building intelligent, tool-equipped agents with memory and reasoning.

A brutally fault-tolerant Mixture-of-Agents (MoA) pipeline built in pure Python. Designed to orchestrate chaotic, round-robin LLM proxy endpoints through a rigorous 4-stage Agentic Workflow (Generate ➔ Cross-Critique ➔ Rebuttal ➔ Judge). Built to eradicate hallucination and guarantee absolute accuracy in complex, multi-step reasoning tasks.

This repository contains the implementation for our Deep NLP course project at Politecnico di Torino. We apply Mixture-of-Adapters (MoA) for RoBERTa and a Variety-Aware Tensor-of-Cues (VAToC) strategy for Mistral-7B (LoRA) to achieve robust sarcasm detection across English varieties on the BESSTIE benchmark.

Multi-model AI debate system. Ask 14 LLMs, get one trusted answer. Adaptive routing, web research on disagreement, adversarial debates, code review with famous engineer personas.

LLM Mixture-of-Agents

Simulation framework for NHS emergency department triage optimization using a Mixture-of-Agents (MoA) architecture — built with SimPy, LangGraph, and FHIR-compliant synthetic data to reduce patient wait times and improve resource utilization.

🤖 An intelligent document Q&A system 📄 leveraging LangChain 🦜, Retrieval-Augmented Generation (RAG) 🧠, and a custom Mixture of Idiots Agents (MoIA) 🤪 approach with OpenAI models 🚀. Ask questions from your documents and receive intelligently synthesized answers! ✨