Awesome AI Agents › Stateful Serverless Frameworks
The project "embodied-agents" by mbodiai is a comprehensive toolkit designed to seamlessly integrate state-of-the-art transformer models into robotics stacks. It aims to simplify the integration of large multi-modal AI models into existing robotic systems with minimal code, providing consistency, reliability, and scalability. The toolkit is structured around three main components: Agents, Data, and Hardware, each further divided into three meta-modalities: Language, Motion, and Sense. This design is inspired by the efficiency of the central nervous system, where each agent type has a specific role: Language Agents return strings, Motor Agents return motions, and Sensory Agents return sensor readings. The project supports both local and remote inference, allowing synchronous or asynchronous execution through various clients like Gradio, httpx, and different large language model (LLM) clients. It includes native support for popular AI models and platforms such as OpenAI, Anthropic, Google Gemini, Ollama, vLLM, and OpenVLA for motor control. Sensory agents support advanced perception models like Depth Anything, YOLO, and Segment Anything 2. Key features include dynamic agent selection via the AutoAgent, extensive API endpoints, and tools for dataset recording on robots. The project also offers practical examples and notebooks for real hardware and simulation environments, facilitating easy experimentation and deployment. The motivation behind the project is to lower the barrier for researchers, hobbyists, and developers to use cutting-edge AI in robotics by addressing challenges like real-time responsiveness on edge devices and the complexity of multi-model integration. Overall, embodied-agents provide a modular, extensible, and observation/action-space agnostic framework that enhances the development and deployment of intelligent robotic systems, promoting data sharing and collaborative advancements in robotics AI.
https://github.com/mbodiai/embodied-agents
LiteLLM is a Python SDK and proxy server that provides a unified OpenAI-compatible interface to call over 100 large language model APIs from multiple providers, featuring consistent output, retry logic, budget controls, and extensive observability integrations.
Flowise is an open-source drag-and-drop platform that enables users to easily build and deploy customized Large Language Model (LLM) application workflows with a user-friendly interface and flexible deployment options.
Parlant is a Conversation Modeling engine that enables precise, consistent, and reliable control over GenAI-driven conversational agents by enforcing structured behavioral guidelines and adapting dynamically to user interactions.
ToolBench is an open platform for training, serving, and evaluating large language models with advanced tool-use capabilities using a large-scale, richly annotated dataset of real-world APIs.
RivetKit is a stateful serverless framework that enables building scalable, real-time, and collaborative applications deployable across multiple platforms including Rivet, Cloudflare Workers, Bun, and Node.js.
LlamaDeploy is an async-first framework that enables seamless deployment, scaling, and productionization of agentic multi-service workflows built with llama_index, facilitating easy transition from development to cloud-based production environments.
Ragbits is a modular framework providing building blocks for rapid development, deployment, and monitoring of scalable and reliable Generative AI applications with flexible LLM integration and advanced document processing capabilities.
Mirascope is a flexible and user-friendly library that provides a unified interface to work with multiple large language model providers, simplifying AI-driven text generation and information extraction tasks.