tools
Enterprise h2oGPTe
Enterprise AI platform for searching private data, generating structured outputs, and running agentic workflows with citations and governance controls.

Enterprise h2oGPTe lets organizations chat with documents and collections, search private knowledge sources, extract structured data, and run agents that perform multi-step tasks such as research, coding, and database access. It supports citation-backed retrieval, multimodal document analysis, model routing, guardrails, and role-based administration.
Organizations can deploy it in public or private cloud, on premises, or in air-gapped environments. A freemium tier is available through the H2O Generative AI app store, while enterprise deployment, support, and managed capabilities require contacting H2O.ai.
Features
- Chat with documents and collections using retrieval-augmented generation
- Run agents with code execution, web search, and tool use
- Generate structured JSON from documents with schema-driven extraction
- Analyze audio, images, flowcharts, and handwritten documents
- Provide citation-backed answers with document references
- Apply guardrails, PII controls, roles, permissions, and audit controls
- Connect through REST APIs and a Python client library
- Deploy on premises, in private or public cloud, or in air-gapped environments
Use cases
- Analyze financial reports, filings, earnings transcripts, and spreadsheets
- Search internal knowledge bases and workplace documents
- Extract contract, compliance, and audit data into structured JSON
- Automate multi-step research and generate citation-backed reports
- Support employees with HR, IT, sales, and marketing information
- Connect AI agents to databases, APIs, and business tools
Pros
Cons
Latest updates
- What's new in h2oGPTe v1.7.x (v1.7.x)
New features and improvements in the h2oGPTe 1.7.x series expand agent control and introduce enterprise-grade governance capabilities.
- v1.7.5 (v1.7.5)
Adds AI Assistants, agent skills, and an LLM registry; rewrites guardrail categories, gives Prompt Guard a new backend, and fixes scheduling and Graph RAG issues.
- What's new in h2oGPTe v1.7.x (v1.7.x)
New features and improvements expand agent control and introduce governance capabilities.
- v1.7.4 (v1.7.4)
Completes tool calling on the OpenAI-compatible API and fixes authentication and Python client issues.
- v1.7.3-post1 (v1.7.3-post1)
Adds Graph RAG, Guardrail Insights, an Agent Tool Server, and admin session management, along with reliability fixes.
Capabilities
- Searches the web — “Agents H2O AI Super Agent: General, Deep Research, Task, Data Science, Web Search, Tool Builder, Agents Builder, MCP Runner, Code” source
- Reads your files — “01 Ingest Upload or connect document sources. Document AI parses the scans, tables, and forms for you.” source
- Acts for you — “Plain English in; a plan, the tool calls, and the result out.” source
- Runs code — “Tools 27 built-in, Python and shell, 16 browser actions; MCP servers (local and remote); A2A protocol” source
- Developer API — “APIs REST API and Python client” source
Get it
Security
Pricing
- Starting price
- Free
- Prices checked
- 2026-10-24
Freemium
Free
- Full access to core functionalities
- Limitations on sharing Collections or Chats
- Thresholds on Collections, Documents, Chats, and LLM usage
- Reduced performance infrastructure
Enterprise / Premium
Price on request
- Upgrade for full sharing and higher limits
- Premium performance infrastructure
- Enterprise deployment options (on-prem, air-gapped, etc.)