Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/cack195/local-rag-chat-assistant
Local AI Assistant using Ollama model (llama3)
https://github.com/cack195/local-rag-chat-assistant
Last synced: 8 days ago
JSON representation
Local AI Assistant using Ollama model (llama3)
- Host: GitHub
- URL: https://github.com/cack195/local-rag-chat-assistant
- Owner: cack195
- Created: 2024-06-18T10:13:38.000Z (7 months ago)
- Default Branch: main
- Last Pushed: 2024-06-18T13:06:14.000Z (7 months ago)
- Last Synced: 2024-11-15T14:19:27.438Z (2 months ago)
- Language: Python
- Size: 23.4 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Local-RAG-Chat-Assistant
Local AI Assistant using flask and Ollama model (llama3)
## Setup
- First fork or clone the project
```sh
git clone https://github.com/cack195/Local-RAG-Chat-Assistant.git
```
### Install Ollama and Download Models
- Follow the [installation guide for Ollama](https://github.com/ollama/ollama?tab=readme-ov-file#macos).
- Next, download the language models (LLMs) you plan to use:
```sh
ollama pull phi3
ollama pull llama3
ollama pull nomic-embed-text
```
### Set Up a Virtual Environment
- Create a Python virtual environment:
```sh
python3 -m venv venv
```
- Activate the virtual environment:
```sh
source venv/bin/activate
```
### Install Required Libraries
Install the necessary Python libraries:
```sh
pip install -r requirements.txt
```
### Start PgVector
Ensure [Docker Desktop](https://docs.docker.com/get-docker/) is installed first.
- Start PgVector using Docker:
```sh
docker run -d \
-e POSTGRES_DB=ai \
-e POSTGRES_USER=ai \
-e POSTGRES_PASSWORD=ai \
-e PGDATA=/var/lib/postgresql/data/pgdata \
-v pgvolume:/var/lib/postgresql/data \
-p 5532:5432 \
--name pgvector \
phidata/pgvector:16
```
### launch the Application
- Run the following command
```sh
python3 app.py
```
- Open [http://127.0.0.1:5000/](http://127.0.0.1:5000/) in your browser to view the local RAG app.
- You can add websites or PDFs and ask questions.