#1818th of 26 in RAG and knowledge
llmware
Python framework for RAG pipelines with a catalog of 300+ local models
- Stars
- 14.8k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
llmware is a Python library that combines a model catalog (300+ models in GGUF, OpenVINO, ONNXRuntime, PyTorch and Hugging Face formats, plus OpenAI, Anthropic and Google APIs) with a RAG pipeline. The pipeline parses mixed document types into libraries, embeds them into a chosen vector DB, and runs text, semantic or hybrid queries. It also ships its own 1-7B BLING, DRAGON and SLIM models for laptop-scale inference.
Who it is for: Python developers building private document Q&A apps for laptops or edge devices
Strengths
- One ModelCatalog interface across GGUF, OpenVINO, ONNXRuntime, PyTorch and cloud APIs
- Ingests pdf, docx, pptx, xlsx, csv, html, wav and images through one add_files call
- Mix and match embedding models and vector DBs per library, such as Milvus or ChromaDB
- Includes post-inference evidence checks against the attached source material
Weaknesses
- Library only; no web UI or server is described in the README
- Needs a text-collection database and vector DB; setup details are not in the excerpt
- RAG examples lean on llmware's own finetuned 1-7B models, so quality on other models is unknown
- No Docker or compose files detected in the repository
What it needs
- GPU optional
- Needs Milvus, ChromaDB
- Models: GGUF, OpenVINO, ONNXRuntime, ONNXRuntime-QNN, PyTorch
Also in RAG and knowledge
See all 26| Rank | Project | Score |
|---|---|---|
| 1 | deeptutorSelf-hosted AI tutor with knowledge bases, guided learning, books and quizzes | 80 out of 100 |
| 2 | Open NotebookSelf-hosted NotebookLM alternative with podcasts and 20+ model providers | 76 out of 100 |
| 3 | LightRAGGraph-based RAG framework with a REST API server and web UI | 76 out of 100 |
| #4 | RAGFlowRAG engine with deep document parsing, template chunking and agentic retrieval | 74 out of 100 |
| #5 | MaxKBEnterprise knowledge-base agent platform with RAG, workflows and MCP tools | 71 out of 100 |