README
¶
pgEdge RAG Server
A simple API server for performing Retrieval-Augmented Generation (RAG) of text based on content from a PostgreSQL database using pgvector.
Features
- Multiple RAG pipelines with configurable embedding and LLM providers
- Hybrid search combining vector similarity and BM25 text matching
- Support for OpenAI, Anthropic, Voyage, and Ollama LLM providers
- Token budget management to control LLM costs
- Optional streaming responses via Server-Sent Events
- TLS/HTTPS support
Quick Start
Prerequisites
- Go 1.22 or later
- PostgreSQL with pgvector extension
- API keys for your chosen LLM providers
Installation
# Clone the repository
git clone https://github.com/pgEdge/pgedge-rag-server.git
cd pgedge-rag-server
# Build the binary
make build
# Run the server
./bin/pgedge-rag-server -config /path/to/config.yaml
Configuration
Create a configuration file (see docs/configuration.md for full reference):
server:
listen_address: "0.0.0.0"
port: 8080
pipelines:
- name: "my-docs"
description: "Search my documentation"
database:
host: "localhost"
port: 5432
database: "mydb"
tables:
- table: "documents"
text_column: "content"
vector_column: "embedding"
embedding_llm:
provider: "openai"
model: "text-embedding-3-small"
rag_llm:
provider: "anthropic"
model: "claude-sonnet-4-20250514"
API Usage
List Available Pipelines
curl http://localhost:8080/v1/pipelines
Query a Pipeline
curl -X POST http://localhost:8080/v1/pipelines/my-docs \
-H "Content-Type: application/json" \
-d '{"query": "How do I configure replication?"}'
Query with Streaming
curl -X POST http://localhost:8080/v1/pipelines/my-docs \
-H "Content-Type: application/json" \
-d '{"query": "How do I configure replication?", "stream": true}'
Development
# Run all checks (format, lint, test, build)
make all
# Run tests only
make test
# Run linter only
make lint
# Format code
make fmt
Documentation
Full documentation is available in the docs/ directory.
License
See LICENCE.md for license information.
Directories
¶
| Path | Synopsis |
|---|---|
|
cmd
|
|
|
pgedge-rag-server
command
|
|
|
internal
|
|
|
bm25
Package bm25 provides BM25 text search scoring.
|
Package bm25 provides BM25 text search scoring. |
|
config
Package config handles configuration loading and validation for the pgEdge RAG Server.
|
Package config handles configuration loading and validation for the pgEdge RAG Server. |
|
database
Package database provides PostgreSQL connectivity and vector search.
|
Package database provides PostgreSQL connectivity and vector search. |
|
llm
Package llm provides interfaces and implementations for LLM providers.
|
Package llm provides interfaces and implementations for LLM providers. |
|
llm/anthropic
Package anthropic provides an Anthropic API client.
|
Package anthropic provides an Anthropic API client. |
|
llm/factory
Package factory provides functions to create LLM providers from configuration.
|
Package factory provides functions to create LLM providers from configuration. |
|
llm/ollama
Package ollama provides an Ollama API client for local LLM inference.
|
Package ollama provides an Ollama API client for local LLM inference. |
|
llm/openai
Package openai provides an OpenAI API client.
|
Package openai provides an OpenAI API client. |
|
llm/voyage
Package voyage provides a Voyage AI embedding client.
|
Package voyage provides a Voyage AI embedding client. |
|
pipeline
Package pipeline provides the RAG pipeline execution and management.
|
Package pipeline provides the RAG pipeline execution and management. |
|
server
Package server provides the HTTP server for the RAG API.
|
Package server provides the HTTP server for the RAG API. |
Click to show internal directories.
Click to hide internal directories.