Awesome AI Agents › Stateful Serverless Frameworks
FireAct is a project focused on advancing the fine-tuning of language agents, as detailed in the publication "FireAct: Toward Language Agent Fine-tuning." The repository provides a comprehensive framework including prompts, demo code, and fine-tuning datasets generated by the authors. It supports defining tools and tasks, generating training data, running experiments, and saving results. The project facilitates the creation of training data in formats compatible with popular fine-tuning frameworks such as Alpaca and GPT, enabling users to prepare datasets for supervised fine-tuning of language models. The repository includes detailed setup instructions, requiring environment variables for OpenAI and SERP API keys, and provides commands for creating a virtual environment and installing dependencies. Users can generate data, perform supervised fine-tuning on models like Llama 2, and run inference with both Llama-based and GPT-based backends. The project also offers a model zoo with multitask models fine-tuned on the Llama family, available on Hugging Face, including LoRA and full model variants. FireAct builds upon and integrates with several existing projects, such as ReAct for generation code, Stanford Alpaca for full model training, Alpaca-LoRA for LoRA training, and a ChatGPT fine-tuning UI for GPT fine-tuning. This integration highlights its role as a versatile and extensible platform for language agent fine-tuning research and development. Overall, FireAct aims to provide researchers and developers with tools and resources to improve the fine-tuning process of language agents, enabling more effective and efficient training and deployment of advanced language models for various tasks.
https://github.com/anchen1011/FireAct
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.