llmware
Basic Information
llmware is an open framework for building LLM-based enterprise applications, with a strong focus on Retrieval-Augmented Generation (RAG) pipelines and multi-model agent workflows. It bundles a Model Catalog of small, specialized models (SLIM, BLING, DRAGON and Industry BERT series) and provides integrated components to ingest, parse, chunk, embed and index documents into libraries. The repo exposes primitives for querying libraries, packaging retrieved evidence into prompts, running local or API-backed models, and orchestrating multi-step analyses with agents and function calls. It targets developers and teams who need to deploy private, cost-effective knowledge-centric LLM applications, offering examples that run on laptops without GPUs as well as deployment patterns for servers and clusters. The project also documents supported vector and text databases, optional native libraries for OCR and audio, and end-to-end examples for contract analysis, chatbots, lecture tools and financial research.
Links
Categorization
App Details
-
$$$$ -
$$$$