Enterprise H2O GPTe: RAG and the Gen AI Ecosystem
- Generative AI
- Enterprise h2oGPTe
- H2O Cascade
This lesson walks through a webinar recap of Enterprise H2O GPTe, H2O's platform for chatting with a company's internal documents without sending data to an external cloud provider. It covers what retrieval augmented generation is and why it matters for grounding answers in real context, then places Enterprise H2O GPTe within the broader H2O generative AI ecosystem alongside the open source H2O GPT and H2O LLM Studio. The lesson closes with a live demo that uploads an EU AI Act document, sets collection level prompts, and queries the text for summaries with page level source citations.
Key takeaways
- Why retrieval augmented generation pairs a retrieval model with a generative model to ground answers in documents
- How to create a collection in Enterprise H2O GPTe and set default prompts that apply to every chat inside it
- The four generation approaches available in a collection, from LLM only to embedded RAG with double queries
- How uploaded documents return answers with page numbers and accuracy scores so sources can be checked
- The difference between using the Enterprise H2O GPTe interface and calling the same features through the Python API
Start owning your AI