Vanilla Agents is a lightweight PyTorch-based framework that delivers modular and extensible implementations of core reinforcement learning agents. It supports algorithms like DQN, Double DQN, PPO, and A2C, with pluggable environment wrappers compatible with OpenAI Gym. Users can configure hyperparameters, log training metrics, save checkpoints, and visualize learning curves. The codebase is organized for clarity, making it ideal for research prototyping, educational use, and benchmarking new ideas in RL.

Agent Script provides a declarative scripting layer over large language models, enabling you to write YAML or JSON scripts that define agent workflows, tool calls, and memory usage. You can plug in OpenAI, local LLMs, or other providers, connect external APIs as tools, and configure long-term memory backends. The framework handles context management, asynchronous execution, and detailed logging out of the box. With minimal code, you can prototype chatbots, RPA workflows, data extraction agents, or custom control loops, making it easy to build, test, and deploy AI-powered automations.

Minimalist Agent provides a bare-bones framework for building AI agents in Python. It leverages LangChain’s agent classes and OpenAI’s API to perform multi-step reasoning, dynamically select tools, and execute functions. You can clone the repository, configure your OpenAI API key, define custom tools or endpoints, and run the CLI script to interact with the agent. The design emphasizes clarity and extensibility, making it easy to study, modify, and extend core agent behaviors for experimentation or teaching.

Embedded BDI provides a full BDI lifecycle engine: it models an agent’s beliefs about its environment, manages evolving desires or goals, selects intentions from a library of plans, and executes behaviors in real time. The framework includes modules for belief base storage, plan library definition, event triggering, and concurrency control tailored for memory-limited microcontrollers. With a simple API, developers annotate beliefs, specify desires, and implement plans in code. Its scheduler handles priority-based intention execution and integrates with hardware interfaces for sensors, actuators, and network communication, making it ideal for autonomous IoT devices, mobile robots, and industrial controllers.

In practice, LlamaSim allows you to define multiple AI-powered agents using the Llama model, set up interaction scenarios, and run controlled simulations. You can customize agent personalities, decision-making logic, and communication channels using simple Python APIs. The framework automatically handles prompt construction, response parsing, and conversation state tracking. It logs all interactions and provides built-in evaluation metrics such as response coherence, task completion rate, and latency. With its plugin architecture, you can integrate external data sources, add custom evaluation functions, or extend agent capabilities. LlamaSim’s lightweight core makes it suitable for local development, CI pipelines, or cloud deployments, enabling replicable research and prototype validation.

MASlite provides a clear API to create agent classes, register behaviors, and handle event-driven messaging between agents. It includes a scheduler to manage agent tasks, environment modeling to simulate interactions, and a plugin system to extend core capabilities. Developers can rapidly prototype multi-agent scenarios in Python by defining agent lifecycle methods, connecting agents via channels, and running simulations in a headless mode or integrating with visualization tools.

SelfDrivingCarSimulator is a lightweight Python framework built on Pygame that offers a 2D driving environment for training autonomous vehicle agents using reinforcement learning. It supports customizable track layouts, configurable sensor models (like LiDAR and camera emulation), real-time visualization, and data logging for performance analysis. Developers can integrate their RL algorithms, adjust physics parameters, and monitor metrics such as speed, collision rate, and reward functions to iterate quickly on self-driving research and educational projects.

DocumentAI-Backend is a lightweight backend framework that automates extraction of text, form fields, and structured data from documents. It offers REST API endpoints for uploading PDFs or images, processes them via Google Document AI with OCR fallback, and returns parsed results in JSON. Built with Python, FastAPI, and Docker, it enables quick integration into existing systems, scalable deployments, and customization through configurable pipelines and middleware.