FreeThinker provides a modular architecture for defining AI agents that can autonomously execute tasks by leveraging large language models, memory modules, and external tools. Developers can configure agents via Python or YAML, plug in custom tools for web search, data processing, or API calls, and utilize built-in planning strategies. The framework handles step-by-step execution, context retention, and result aggregation so agents can operate hands-free on research, automation, or decision-support workflows.

Agentic Workflow is a declarative framework enabling developers to define complex AI workflows by chaining multiple LLM-based agents, each with customizable roles, prompts, and execution logic. It provides built-in support for task orchestration, state management, error handling, and plugin integrations, allowing seamless interaction between agents and external tools. The library uses Python and YAML-based configurations to abstract agent definitions, supports asynchronous execution flows, and offers extensibility through custom connectors and plugins. As an open-source project, it includes detailed examples, templates, and documentation to help teams accelerate development and maintain complex AI agent ecosystems.

MARL-DPP is an open-source framework enabling multi-agent reinforcement learning (MARL) with enforced diversity through Determinantal Point Processes (DPP). Traditional MARL approaches often suffer from policy convergence to similar behaviors; MARL-DPP addresses this by incorporating DPP-based measures to encourage agents to maintain diverse action distributions. The toolkit provides modular code for embedding DPP in training objectives, sampling policies, and managing exploration. It includes ready-to-use integration with standard OpenAI Gym environments and the Multi-Agent Particle Environment (MPE), along with utilities for hyperparameter management, logging, and visualization of diversity metrics. Researchers can evaluate the impact of diversity constraints on cooperative tasks, resource allocation, and competitive games. The extensible design supports custom environments and advanced algorithms, facilitating exploration of novel MARL-DPP variants.