LLAMARA - Large Language Assistant for Model-Augmented Retrieval and Analysis - is an LLM-based assistant for information retrieval from a provided knowledge base. It aims at supporting researchers working with scientific papers, whitepapers and documentation, as well as possibly serving research findings in an accessible way to the public.
This repository contains a Docker Compose file and a configuration template to deploy the LLAMARA distribution Docker container.
This application requires an OIDC authentication provider to be set up. We recommend to use Keycloak as OIDC provider.
You need to add the microprofile-jwt and profile scopes for the Quarkus client, e.g. for Keycloak see Keycloak Server Documentation.
Provide the required environment variables in a .env file:
QUARKUS_OIDC_CREDENTIALS_SECRET=
POSTGRES_PASSWORD=
MINIO_ROOT_PASSWORD=
MINIO_ACCESS_KEY=
MINIO_SECRET_KEY=
REDIS_PASSWORD=
QDRANT_API_KEY=
OPENAI_API_KEY=Use application.yaml.sample to create a application.yaml in the config dir that provides the required configuration.
Refer to llamara-backend to learn about the supported configuration properties.
Once you have completed the configuration step, start LLAMARA through Docker Compose:
docker compose up