https://github.com/natan1995/llm_docker_service
MIcroservice that was created to connect a pre trained LLM to a different databases
https://github.com/natan1995/llm_docker_service
docker docker-compose fastapi llm-agents postgresql python3
Last synced: 3 months ago
JSON representation
MIcroservice that was created to connect a pre trained LLM to a different databases
- Host: GitHub
- URL: https://github.com/natan1995/llm_docker_service
- Owner: Natan1995
- Created: 2025-02-09T15:04:32.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-16T21:33:02.000Z (over 1 year ago)
- Last Synced: 2025-02-16T22:24:31.733Z (over 1 year ago)
- Topics: docker, docker-compose, fastapi, llm-agents, postgresql, python3
- Language: Python
- Homepage:
- Size: 7.81 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Lienzu LLM
Lienzu LLM is an open source project that seeks to integrate data sources from a company without exposing sensitive data from them; for this purpose, an isolated ecosystem made up of four elements is developed:
- A local microservice developed with FastAPI.
- A database built on postgres SQL.
- An LLM that understands the needs of non-technical users. Converting prompts into SQL queries.
- A second LLM that returns the best possible answer to the query made by the initial user.
## Installation
Deploy the Docker images with docker compose, open your terminal and execute this commands.
```bash
mkdir ./init.sql
wget -O ./init.sql/init_users.sql https://raw.githubusercontent.com/Natan1995/llm_docker_service/refs/heads/main/init.sql/init_users.sql
wget -O docker-compose.yml https://raw.githubusercontent.com/Natan1995/llm_docker_service/refs/heads/main/docker-compose.yml
docker-compose up
```
## Usage/Examples
In a browser of your preference, you can get the database information with these routes:
```javascript
Get all users from database:
http://localhost:8000/users
Get all products from database:
http://localhost:8000/products
Get all sales from database:
http://localhost:8000/sales
Custom queries for databases:
http://localhost:8000/query/?query=SELECT%20*%20FROM%20users
```
## Authors
- [@Natan1995](https://github.com/Natan1995)