Practical overview
When it fits
Use it when you need a documented starting point for a self-hosted document-chat workflow built with Spring Boot, Spring AI and PostgreSQL vector search, or want to study how an Ollama model path is configured.
Who it is for
- Developers exploring Spring AI retrieval-augmented generation.
- Teams considering a self-hosted document question-answering application.
- Java developers adapting a document upload, vector search and chat workflow.
Documented uses
- Upload documents to a library and make their content searchable through embeddings.
- Ask questions and retrieve nearby document content as context for AI-generated answers.
- Follow a source-document link to research the material behind an answer.
Requirements & limits
The repository is public and specifies an Apache-2.0 license. Its documented setup is for self-hosting; the project lists software prerequisites rather than a hosted signup process.
- The main workflow retrieves document content by vector similarity before generating an answer, and provides a source-document link.
- PostgreSQL with Vector Extension 0.5.1 or newer is listed as a setup prerequisite.
- Ollama requires a build with useOllama=true and startup with the ollama profile; CPU performance may be an issue.
- The frontend uses Angular, Angular Material and TypeScript; backend services use Spring Boot, Spring AI, JPA and Liquibase.
- A Helm setup for minikube is documented alongside the local development prerequisites.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up the documented local development prerequisites
Prepare the listed database and development tools needed to work with the application.
Before you start
- PostgreSQL with Vector Extension 0.5.1 or newer
- Java 25 or newer
- Gradle 9.1 or newer
- Node.js 22.15.x or newer
- npm 10.9.x or newer
Steps
Provide a PostgreSQL database with Vector Extension 0.5.1 or newer.
Install Java 25 or newer and Gradle 9.1 or newer.
Install Node.js 22.15.x or newer, npm 10.9.x or newer, and Angular CLI 20 or newer.
Expected outcome
The documented database and development prerequisites are in place for working with the project.
Guide limitations
- These instructions cover prerequisites only; they do not document the subsequent build or application run steps.
Configure the documented Ollama application mode
Select the repository's Ollama build and application profile for locally run models.
Before you start
- Ollama available to run a local model
- The project build environment
Steps
Build the application with the Gradle property useOllama=true to include Ollama dependencies.
Start the application with the ollama profile to switch on the Ollama configuration and features.
Expected outcome
The project is configured to use its Ollama model path.
Guide limitations
- The repository notes that CPU performance can be an issue.
- The instructions do not identify a tested model or guarantee a deployment outcome.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 13 October 2026
- Original website title
- Angular2Guy/AIDocumentLibraryChat · Official homepage title · retained in its source language