aquiles-rag 0.2.0

Aquiles-RAG is a high-performance Retrieval-Augmented Generation (RAG) solution built on Redis. It offers a high-level interface through FastAPI REST APIs

Aquiles-RAG

Description

Aquiles-RAG is a high-performance Retrieval-Augmented Generation (RAG) solution built on Redis. It offers a high-level interface through FastAPI REST APIs to:

• Create RAG indexes in Redis.
• Send raw text alongside its embeddings (the client must chunk the text and compute embeddings before submission).
• Query the index to retrieve the most relevant chunks.

Features

• Optimized Performance: Uses Redis as a vector search engine.
• Simple API: Endpoints for index creation, insertion, and querying.
• Extensible: Basic implementation ready for enhancements and integration into ML pipelines.

High-Level Architecture

Here's a diagram illustrating how Aquiles-RAG connects clients to Redis using an asynchronous FastAPI server:

Usage

Create Index

curl -X POST http://localhost:5500/create/index \
     -H 'Content-Type: application/json' \
     -d '{"indexname": "my_index"}'

Send RAG

curl -X POST http://localhost:5500/rag/create \
     -H 'Content-Type: application/json' \
     -d '{
           "index": "my_index",
           "raw_text": "Full text goes here...",
           "embeddings": [0.12, 0.34, ...]
         }'

Query RAG

curl -X POST http://localhost:5500/rag/query-rag \
     -H 'Content-Type: application/json' \
     -d '{
           "index": "my_index",
           "embeddings": [0.56, 0.78, ...],
           "top_k": 5
         }'

Command-Line Interface (CLI)

Usage Examples

# Save configuration
aquiles-rag configs --local True --host redis.local --port 6380

# Start server
aquiles-rag serve