run-llama/llama_index
LlamaIndex is a leading data framework that enables building LLM-powered applications by providing tools for data ingestion, structuring, and advanced querying to augment large language models with private and external data.
Awesome AI Agents › Data Integration and Specialized Solutions
getzep/graphiti
Graphiti is an advanced framework designed for building and querying real-time, temporally-aware knowledge graphs tailored specifically for AI agents operating in dynamic environments. Unlike traditional retrieval-augmented generation (RAG) methods that rely on batch processing and static data summarization, Graphiti continuously integrates user interactions, structured and unstructured enterprise data, and external information into a coherent, queryable graph. This allows for incremental data updates, efficient retrieval, and precise historical queries without the need for complete graph recomputation, making it ideal for interactive, context-aware AI applications. The framework supports a bi-temporal data model that explicitly tracks both event occurrence and ingestion times, enabling accurate point-in-time queries and temporal edge invalidation to handle contradictions. Graphiti combines semantic embeddings, keyword (BM25), and graph traversal methods to provide low-latency hybrid retrieval without relying on large language model (LLM) summarization. It also offers flexible ontology creation with custom entity definitions through developer-friendly Pydantic models, ensuring adaptability to various use cases. Graphiti is scalable and optimized for large datasets, supporting parallel processing suitable for enterprise environments. It powers the core memory layer of Zep, an AI agent memory system, and has been demonstrated as state-of-the-art in agent memory through published research. The project is open-source and encourages community involvement. Installation requires Python 3.10+, Neo4j 5.26+ as the embeddings storage backend, and an OpenAI API key for LLM inference and embedding, with optional support for other LLM providers like Google Gemini, Anthropic, and Groq. The project includes comprehensive documentation, examples, and a quickstart guide to help users connect to Neo4j, initialize indices, add data episodes, and perform complex searches using hybrid and graph-based methods. Overall, Graphiti addresses the challenges of managing dynamic and frequently updated datasets, enabling real-time interaction and precise historical data queries for AI agents and other applications requiring sophisticated knowledge graph capabilities.
https://github.com/getzep/graphiti
Also in Data Integration and Specialized Solutions
pingcap/tidb
TiDB is an open-source, cloud-native, distributed SQL database offering ACID guarantees, horizontal scalability, high availability, HTAP capabilities, and MySQL compatibility.
dolthub/dolt
Dolt is a SQL database with Git-like features, enabling full version control over data, including branching, merging, and diffing of data.
vectordotdev/vector
Vector is a high-performance, end-to-end observability data pipeline designed to collect, transform, and route logs, metrics, and traces.
cube-js/cube
Cube Core is an open-source semantic layer that enables AI, business intelligence, and embedded analytics by providing a flexible, API-driven platform supporting multiple SQL data sources and real-time analytics.
llmware-ai/llmware
llmware is a unified framework for building enterprise Retrieval-Augmented Generation (RAG) pipelines using small, specialized language models integrated with secure knowledge sources for efficient AI applications.
cocoindex-io/cocoindex
CoCoIndex is an ultra-performant data transformation framework for AI, specializing in incremental processing for data indexing and real-time semantic search with Python and Rust.
electric-sql/electric
Electric is a Postgres sync engine that provides real-time data synchronization, partial replication, and data delivery for modern applications and AI agents.