Awesome AI Agents › Stateful Serverless Frameworks
Parlant is a sophisticated Conversation Modeling engine designed to enhance and control interactions with Generative AI (GenAI) agents, ensuring power, precision, and consistency in conversational experiences. It addresses the common challenge faced by developers and businesses in making AI agents behave according to specific business rules and user expectations, especially in customer-facing scenarios. Unlike traditional flow engines that restrict user interactions to predefined paths or free-form prompt engineering that often leads to inconsistent responses, Parlant dynamically adapts to natural user interactions while strictly enforcing a structured set of behavioral guidelines called Conversation Models. A Conversation Model in Parlant is a domain-specific framework comprising principles, actions, objectives, and terminology that guide an AI agent's behavior during conversations. This approach is particularly valuable in regulated and sensitive industries such as financial services, healthcare, legal assistance, compliance, brand-sensitive customer service, and personal advocacy, where precision and adherence to protocols are critical. Parlant empowers developers and data scientists to create custom conversational agents quickly by defining clear behavioral guidelines that the agents must follow. It supports attaching tools with contextual usage instructions, managing domain-specific glossaries for consistent term interpretation, and incorporating customer-specific data for personalized interactions. The engine operates by reacting to session triggers, loading relevant domain terminology, matching guidelines, invoking tools as needed, and composing guided messages. It also includes mechanisms for self-critique to ensure responses align with intended behaviors. The platform offers features like session management, incoherence and contradiction detection in guidelines, content filtering, jailbreak protection, and an integrated sandbox UI for testing. It provides native API clients in Python and TypeScript and supports all major LLM providers, including OpenAI and Gemini. Parlant is ideal for projects requiring continuous behavioral refinement, coherent guideline management with version tracking, and enhanced conversational user experience control. Its unique focus on managed guidelines delivers reliability, explainability, and maintainability in AI-driven conversations.
https://github.com/emcie-co/parlant
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.
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.
Strands Agents Tools is a versatile toolkit that equips AI agents with powerful capabilities for file operations, system interaction, API communication, multimedia processing, memory management, and advanced reasoning to build intelligent applications efficiently.