This server is designed to connect to multiple Retrieval-Augmented Generation (RAG) systems using different embedding models for the ingestion pipeline. Setup and Run
To set up and run the server, follow these steps:
-
pip install -r requirements.txt
-
python main.py
You can connect multiple RAG systems and use different embedding models for each. Below are examples of configurations for LanceDB and MongoDB.
{
"provider": "LanceDB",
"provider_config": {
"embedding": {
"provider": "OpenAI",
"model_name": "text-embedding-3-small"
},
"chunk_size": 512,
"overlapping": 200,
"worker": 2,
"similarity_top_k": 2,
"rag": {
"loc": "dev"
}
}
} {
"provider": "MongoDB",
"provider_config": {
"embedding": {
"provider": "OpenAI",
"model_name": "text-embedding-3-small"
},
"chunk_size": 512,
"overlapping": 200,
"worker": 2,
"similarity_top_k": 2,
"rag": {
"uri": "mongodb+srv://username:[email protected]/",
"db": "movies",
"collection_name": "movies_records"
}
}
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent