Awesome AI Agents › AI Workflow Orchestrators

EmbeddedLLM/JamAIBase

⭐ 1103 Python added to this list on 2025-04-19 repository created 2024-05-30

JamAI Base is an open-source Retrieval-Augmented Generation (RAG) backend platform designed to facilitate the development and management of AI applications through an intuitive, spreadsheet-like user interface and a simple REST API. It integrates an embedded SQLite database and LanceDB, an embedded vector database, to provide managed memory and RAG capabilities. The platform supports built-in large language models (LLMs), vector embeddings, and reranker orchestration, enabling users to chain cells into powerful pipelines, experiment with prompts and models, and evaluate LLM responses in real-time collaboratively. The platform offers several specialized table types to enhance AI workflows: Generative Tables dynamically generate data using LLMs and provide built-in REST API endpoints; Action Tables enable real-time interaction between frontend applications and the LLM backend, automating backend management and supporting complex workflow orchestration; Knowledge Tables serve as structured repositories for data and documents, enriching the contextual understanding of LLMs and supporting other generative tables; Chat Tables simplify the creation and management of intelligent chatbots with context-aware interactions and seamless integration with RAG. JamAI Base leverages LanceDB for efficient management and querying of large-scale multi-modal data, ensuring scalability and optimal performance. Its declarative paradigm allows users to focus on defining desired outcomes rather than procedural implementation, simplifying development and making advanced AI capabilities accessible to users with varying technical expertise. Key benefits include ease of use through a natural language prompt-driven interface, scalability via a serverless design, flexibility to support any LLM including OpenAI GPT-4, Anthropic Claude 3, and Meta Llama3, and innovative RAG techniques such as query rewriting, hybrid search and reranking, structured content management, adaptive chunking, and multi-lingual embeddings. The platform supports both cloud-based and self-hosted deployment options, backed by comprehensive documentation, community support, and example applications to help users get started quickly.

https://github.com/EmbeddedLLM/JamAIBase

action-tablesadaptive-chunkingagentsaiai-agents-frameworkai-applicationsbaasbackend-as-a-servicechat-tableschatbotchatbot-developmentchatgptdeclarative-paradigmembedded-databasegenerative-tableshybrid-searchintelligent-spreadsheetknowledge-tableslancedblarge-language-modelsllama3-1llmllm-opsmanaged-memorymulti-lingual-embeddingsmulti-modal-datanatural-language-promptsopen-sourceorchestrationpythonquery-rewritingragreranker-orchestrationrest-apiretrieval-augmented-generationscalabilityserverlessspreadsheetspreadsheet-like-uisqlitesveltevector-databasevector-embeddingsworkflow