# h2oGPT

> Apache 2.0 open-source project for querying and summarizing documents or chatting with local private GPT LLMs. Supports Ollama, Mixtral, llama.cpp with persistent databases (Chroma, Weaviate, FAISS) and accurate embeddings.

- **URL:** https://github.com/h2oai/h2ogpt
- **Category:** LLM Frameworks
- **Tags:** Open Source, privacy, local-llm
- **Updated:** 2026-03-11 13:24
- **Canonical page:** https://vectordb.ever.works/items/h2ogpt

## Details

## Overview

h2oGPT is an Apache 2.0 licensed open-source project that enables private document Q&A and chat with local LLMs. It provides 100% private, offline-capable AI interactions with document understanding.

## Features

- **Multiple LLM Backends**: Supports Ollama, Mixtral, llama.cpp, and more
- **Document Processing**: Query and summarize documents privately
- **Persistent Databases**: Integration with Chroma, Weaviate, or in-memory FAISS
- **Quality Embeddings**: Uses instructor-large, all-MiniLM-L6-v2, and others
- **100% Private**: All processing happens locally
- **Apache 2.0**: Permissive open-source license
- **Web UI**: User-friendly interface included
- **API Access**: Programmatic access to capabilities

## Use Cases

- Private document analysis
- Offline RAG applications
- Enterprise document Q&A without data leaving premises
- Healthcare and legal document processing
- Research with sensitive data

## Integration

Works with multiple vector databases and embedding models. Compatible with various local LLM providers.

## Deployment

Can be deployed on-premises or in private cloud environments. Demo available at gpt.h2o.ai.

## Pricing

Free and open-source under Apache 2.0 license.

---

_This Markdown mirror is generated by Ever Works for AI agents. The canonical HTML page is at https://vectordb.ever.works/items/h2ogpt._