Verba

Verba is a retrieval-augmented generation interface and chatbot that uses Weaviate to provide factual answers based on private datasets. It functions as a vector database knowledge base, combining a hybrid search engine with an orchestration interface to connect various large language model providers and embedding services.

The system differentiates itself through a RAG pipeline manager for adjusting text chunking rules and retrieval settings, alongside a 3D vector space visualization tool for analyzing the spatial organization and clustering of high-dimensional embeddings. It employs a modular provider system that allows for swapping between different local and cloud text generation and embedding services.

The platform covers multi-modal data ingestion, processing unstructured documents, audio transcriptions, web crawls, and version control repositories into a searchable knowledge base. Its retrieval capabilities combine semantic and keyword search to extract relevant context from vector stores, utilizing configurable text chunking to optimize retrieval precision.

Features

Retrieval-Augmented Generation - Combines large language models with vector databases to generate factual responses based on specific external datasets.

Private Knowledge Agents - Provides a conversational interface specialized in querying private document sets for grounded, factual answers.

Private Dataset Insight Extraction - Enables the extraction of insights and resolution of specific questions from private datasets using RAG.

LLM Orchestrators - Provides a configuration layer to manage connections between diverse LLM providers and embedding services.

Text Chunks - Segments large documents into smaller pieces using token or semantic rules to optimize retrieval precision.

Retrieval Augmented Generation Pipelines - Allows users to adjust settings for the retrieval-augmented generation process to improve data discovery and answer quality.

Vector Databases - Uses a vector database to store high-dimensional embeddings of unstructured documents and audio for semantic retrieval.

Private Data Querying - Provides a conversational interface to retrieve factual answers from private vector databases.

Data Ingestion - Implements pipelines for loading and parsing PDFs, audio, and web content for AI consumption.

Hybrid Search Engines - Combines keyword and semantic search to find the most relevant information within a vector store.

Hybrid Retrieval - Combines keyword matching and vector similarity to improve the accuracy of retrieved context.

Vector Database Integrations - Integrates with local or cloud-hosted vector database instances to store and retrieve high-dimensional embeddings.

Vector Search - Converts text into high-dimensional embeddings to find documents based on conceptual meaning.

Knowledge Base Retrieval - Retrieves relevant data chunks from a knowledge base to provide grounded context for AI-generated answers.

Pipeline Management Interfaces - Ships a dedicated workspace for adjusting text chunking rules and retrieval settings to optimize answer accuracy.

Model Provider Integrations - Implements a provider-based system to swap between different local and cloud text generation and embedding services.

RAG Chatbot Interfaces - Provides a specialized chat interface that generates factual responses based on retrieved private datasets via Weaviate.

Three-Dimensional Visualizations - Renders high-dimensional embeddings into a 3D coordinate system for spatial analysis of data clusters.

Document Processing Tools - Processes various file formats, including PDFs and plain text, to make raw content searchable for chatbots.

Multi-Format Document Ingestion - Processes audio transcriptions, web crawls, and documents into a standardized format for vector storage.

Embedding Provider Configurations - Provides mechanisms to configure connections to local or cloud services for text generation and vector embeddings.

Embedding Visualizations - Includes a 3D vector space visualization tool to analyze the spatial organization and clustering of embeddings.

Private Knowledge Base Imports - Enables the import of files, directories, or specific URLs to build a private, searchable knowledge base.

RAG Frameworks - Out-of-the-box application for implementing RAG workflows.

Retrieval Augmented Generation - Chatbot interface powered by vector database retrieval.

RAG Frameworks - End-to-end RAG application interface for exploring datasets.

weaviateVerbaArchived

Features

Star history