Multi-Agent-Seller provides a modular environment for simulating e-commerce negotiations using AI agents. It includes pre-built buyer and seller agents with customizable negotiation strategies, such as dynamic pricing, time-based concessions, and utility-based decision-making. Users can define custom protocols, message formats, and market conditions. The framework handles session management, offer tracking, and result logging with built-in visualization tools for analyzing agent interactions. It integrates easily with machine learning libraries for strategy development, enabling experimentation with reinforcement learning or rule-based agents. Its extensible architecture allows adding new agent types, negotiation rules, and visualization plugins. Multi-Agent-Seller is ideal for testing multi-agent algorithms, studying negotiation behaviors, and teaching concepts in AI and e-commerce domains.

VMAS is a comprehensive toolkit for building and training multi-agent systems using deep reinforcement learning. It supports GPU-based parallel simulation of hundreds of environment instances, enabling high-throughput data collection and scalable training. VMAS includes implementations of popular MARL algorithms like PPO, MADDPG, QMIX, and COMA, along with modular policy and environment interfaces for rapid prototyping. The framework facilitates centralized training with decentralized execution (CTDE), offers customizable reward shaping, observation spaces, and callback hooks for logging and visualization. With its modular design, VMAS seamlessly integrates with PyTorch models and external environments, making it ideal for research in cooperative, competitive, and mixed-motive tasks across robotics, traffic control, resource allocation, and game AI scenarios.

AI-Agentic Machine Translation is an open-source framework designed for research and development in machine translation. It orchestrates three core agents—a generator, an evaluator, and a refiner—to collaboratively produce, assess, and refine translations. Built on PyTorch and transformer models, the system supports supervised pre-training, reinforcement learning optimization, and configurable agent policies. Users can benchmark on standard datasets, track BLEU scores, and extend the pipeline with custom agents or reward functions to explore agentic collaboration in translation tasks.

MAPF_G2RL is an open-source research framework that bridges graph theory and deep reinforcement learning to tackle the multi-agent path finding (MAPF) problem. It encodes nodes and edges into vector representations, defines spatial and collision-aware reward functions, and supports various RL algorithms such as DQN, PPO, and A2C. The framework automates scenario creation by generating random graphs or importing real-world maps, and orchestrates training loops that optimize policies for multiple agents simultaneously. After learning, agents are evaluated in simulated environments to measure path optimality, makespan, and success rates. Its modular design allows researchers to extend core components, integrate new MARL techniques, and benchmark against classical solvers.

Multi-Agent Systems provides a comprehensive toolkit for creating, controlling, and observing interactions among autonomous agents. Developers can define agent classes with custom decision-making logic, set up complex environments with configurable resources and rules, and implement communication channels for information exchange. The framework supports synchronous and asynchronous scheduling, event-driven behaviors, and integrates logging for performance metrics. Users can extend core modules or integrate external AI models to enhance agent intelligence. Visualization tools render simulations in real-time or post-process, helping analyze emergent behaviors and optimize system parameters. From academic research to prototype distributed applications, Multi-Agent Systems simplifies end-to-end multi-agent simulations.

Multi-Agent-Blueprint is a comprehensive open-source codebase for building and orchestrating multiple AI-driven agents that collaborate to address complex tasks. At its core, it offers a modular system for defining distinct agent roles—such as researchers, analysts, and executors—each with dedicated memory stores and prompt templates. The framework integrates seamlessly with large language models, external knowledge APIs, and custom tools, enabling dynamic task delegation and iterative feedback loops between agents. It also includes built-in logging and monitoring to track agent interactions and outputs. With customizable workflows and interchangeable components, developers and researchers can rapidly prototype multi-agent pipelines for applications like content generation, data analysis, product development, or automated customer support.