SmartRAG 3.2.0

.NET Standard 2.1 This package targets .NET Standard 2.1. The package is compatible with this framework or higher.
There is a newer version of this package available.
See the version list below for details. 
dotnet add package SmartRAG --version 3.2.0
                    


                    

                
NuGet\Install-Package SmartRAG -Version 3.2.0
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package. 
<PackageReference Include="SmartRAG" Version="3.2.0" />
For projects that support PackageReference, copy this XML node into the project file to reference the package. 
<PackageVersion Include="SmartRAG" Version="3.2.0" />
                    


                            Directory.Packages.props
                        

                    

                
<PackageReference Include="SmartRAG" />
                    


                            Project file
For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package. 
paket add SmartRAG --version 3.2.0
                    


                    

                
#r "nuget: SmartRAG, 3.2.0"
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package. 
#:package SmartRAG@3.2.0
#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package. 
#addin nuget:?package=SmartRAG&version=3.2.0
                    


                            Install as a Cake Addin
                        

                    

                
#tool nuget:?package=SmartRAG&version=3.2.0
                    


                            Install as a Cake Tool

SmartRAG

Multi-Database RAG Library for .NET
Ask questions about your data in natural language

Description

SmartRAG is a comprehensive Retrieval-Augmented Generation (RAG) library that enables you to query multiple databases, documents, images, and audio files using natural language. Transform your data into intelligent conversations with a single, unified API.

Key Capabilities

  • Multi-Database RAG - Query SQL Server, MySQL, PostgreSQL, SQLite together
  • Multi-Modal Intelligence - PDF, Word, Excel, Images (OCR), Audio (Whisper.net)
  • On-Premise Ready - 100% local with Ollama, LM Studio, Whisper.net
  • Conversation History - Built-in automatic context management
  • Universal AI Support - OpenAI, Anthropic, Gemini, Azure, Custom APIs
  • Enterprise Storage - Qdrant, Redis, SQLite, FileSystem, In-Memory

Getting Started

Prerequisites

  • .NET Standard 2.1 or higher
  • AI Provider API key (OpenAI, Anthropic, etc.) or local AI setup (Ollama)
  • Database connections (SQL Server, MySQL, PostgreSQL, SQLite)

Installation

dotnet add package SmartRAG

Basic Setup

// Program.cs
builder.Services.AddSmartRAG(builder.Configuration);

Configuration

Add database connections to your appsettings.json:

{
  "SmartRAG": {
    "DatabaseConnections": [
      {
        "Name": "Sales",
        "ConnectionString": "Server=localhost;Database=Sales;...",
        "DatabaseType": "SqlServer"
      }
    ]
  }
}

Usage

Upload Documents

// Upload document
var document = await documentService.UploadDocumentAsync(
    fileStream, fileName, contentType, "user-123"
);

Query Across Multiple Data Sources

// Query across databases and documents
var response = await documentSearchService.QueryIntelligenceAsync(
    "Show me all customers who made purchases over $10,000 in the last quarter, their payment history, and any complaints or feedback they provided"
);
// → AI automatically queries SQL Server (orders), MySQL (payments), PostgreSQL (customer data), 
//   analyzes uploaded PDF contracts, OCR-scanned invoices, and transcribed call recordings

Multi-Database Queries

// Direct multi-database query coordination
var multiDbResponse = await multiDbCoordinator.QueryMultipleDatabasesAsync(
    "Calculate total revenue by customer across all databases", 
    maxResults: 10
);
// → AI analyzes query intent, generates SQL for each database,
//   executes queries, and merges results intelligently

Real-World Use Cases

Banking - Customer Financial Profile

var answer = await documentSearchService.QueryIntelligenceAsync(
    "Which customers have overdue payments and what's their total outstanding balance?"
);
// → Queries Customer DB, Payment DB, Account DB and combines results
// → Provides comprehensive financial risk assessment for credit decisions

Healthcare - Patient Care Management

var answer = await documentSearchService.QueryIntelligenceAsync(
    "Show me all patients with diabetes who haven't had their HbA1c checked in 6 months"
);
// → Combines Patient DB, Lab Results DB, Appointment DB and identifies at-risk patients
// → Ensures preventive care compliance and reduces complications

Inventory - Supply Chain Optimization

var answer = await documentSearchService.QueryIntelligenceAsync(
    "Which products are running low on stock and which suppliers can restock them fastest?"
);
// → Analyzes Inventory DB, Supplier DB, Order History DB and provides restocking recommendations
// → Prevents stockouts and optimizes supply chain efficiency

What Makes SmartRAG Special?

  • Only .NET library with native multi-database RAG capabilities
  • Automatic schema detection across different database types
  • 100% local processing with Ollama and Whisper.net
  • Enterprise-ready with comprehensive error handling and logging
  • Cross-database queries without manual SQL writing
  • Multi-modal intelligence combining documents, databases, and AI

Supported Data Sources

Databases: SQL Server, MySQL, PostgreSQL, SQLite
Documents: PDF, Word, Excel, PowerPoint, Images, Audio
AI Models: OpenAI, Anthropic, Ollama (local), LM Studio
Vector Stores: Qdrant, Redis, SQLite, InMemory

Comparison with Other RAG Libraries

Feature SmartRAG LM-Kit.NET Semantic Kernel LangChain.NET
Library Owner Barış Yerlikaya LM-Kit (France) Microsoft LangChain Community
Description Multi-database RAG library for .NET Generative AI SDK for .NET applications AI orchestration framework .NET port of LangChain framework
RAG Support ✅ ✅ ✅ ✅
Memory Management ✅ ✅ ✅ ✅
Vector Stores ✅ ✅ ✅ ✅
AI Model Integration ✅ ✅ ✅ ✅
Plugin/Extension System ✅ ✅ ✅ ✅
Multi-Modal ✅ ✅ ❌ ❌
Local AI ✅ ✅ ❌ ❌
Audio ✅ ✅ ❌ ❌
OCR ✅ ✅ ❌ ❌
On-Premise ✅ ✅ ❌ ❌
Fallback Providers* ✅ ❌ ❌ ❌
Retry Policies* ✅ ❌ ❌ ❌
Batch Embeddings* ✅ ❌ ❌ ❌
Hybrid Search* ✅ ❌ ❌ ❌
Session Management* ✅ ❌ ❌ ❌
Cross-DB JOIN* ✅ ❌ ❌ ❌
Multi-DB RAG* ✅ ❌ ❌ ❌
Databases* ✅ ❌ ❌ ❌

SmartRAG Exclusive Features (*):

  • Fallback Providers: Automatic failover to backup AI providers when primary fails
  • Retry Policies: Configurable retry with FixedDelay, LinearBackoff, ExponentialBackoff
  • Batch Embeddings: Efficient batch processing for multiple texts simultaneously
  • Hybrid Search: Semantic + keyword hybrid algorithm (80% semantic, 20% keyword)
  • Session Management: Persistent conversation continuity across app restarts
  • Cross-DB JOIN: AI-powered intelligent joins across different databases
  • Multi-DB RAG: Native multi-database query coordination
  • Databases: Native support for SQL Server, MySQL, PostgreSQL, SQLite

Additional Documentation

Feedback

License

MIT License - see LICENSE for details.

Built with ❤️ by Barış Yerlikaya

Made in Turkey 🇹🇷 | Contact | LinkedIn | Website

Product Compatible and additional computed target framework versions.
.NET net5.0 was computed.  net5.0-windows was computed.  net6.0 was computed.  net6.0-android was computed.  net6.0-ios was computed.  net6.0-maccatalyst was computed.  net6.0-macos was computed.  net6.0-tvos was computed.  net6.0-windows was computed.  net7.0 was computed.  net7.0-android was computed.  net7.0-ios was computed.  net7.0-maccatalyst was computed.  net7.0-macos was computed.  net7.0-tvos was computed.  net7.0-windows was computed.  net8.0 was computed.  net8.0-android was computed.  net8.0-browser was computed.  net8.0-ios was computed.  net8.0-maccatalyst was computed.  net8.0-macos was computed.  net8.0-tvos was computed.  net8.0-windows was computed.  net9.0 was computed.  net9.0-android was computed.  net9.0-browser was computed.  net9.0-ios was computed.  net9.0-maccatalyst was computed.  net9.0-macos was computed.  net9.0-tvos was computed.  net9.0-windows was computed.  net10.0 was computed.  net10.0-android was computed.  net10.0-browser was computed.  net10.0-ios was computed.  net10.0-maccatalyst was computed.  net10.0-macos was computed.  net10.0-tvos was computed.  net10.0-windows was computed. 
.NET Core netcoreapp3.0 was computed.  netcoreapp3.1 was computed. 
.NET Standard netstandard2.1 is compatible. 
MonoAndroid monoandroid was computed. 
MonoMac monomac was computed. 
MonoTouch monotouch was computed. 
Tizen tizen60 was computed. 
Xamarin.iOS xamarinios was computed. 
Xamarin.Mac xamarinmac was computed. 
Xamarin.TVOS xamarintvos was computed. 
Xamarin.WatchOS xamarinwatchos was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages

This package is not used by any NuGet packages.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
3.3.0 720 12/1/2025
3.2.0 329 11/27/2025
3.1.0 317 11/13/2025
3.0.3 422 11/6/2025
3.0.2 309 10/23/2025
3.0.1 377 10/22/2025
3.0.0 367 10/22/2025
2.3.1 229 10/8/2025
2.3.0 351 9/16/2025
2.2.0 387 9/15/2025
2.1.0 155 9/5/2025
2.0.0 317 8/27/2025
1.1.0 174 8/22/2025
1.0.3 189 8/20/2025
1.0.2 192 8/19/2025
1.0.1 163 8/17/2025
1.0.0 124 8/15/2025

Release 3.2.0: Architectural Refactoring & Modular Design (MINOR Release - 2025-11-27)

ARCHITECTURAL REFACTORING:
- Strategy Pattern Implementation: ISqlDialectStrategy, IScoringStrategy, IFileParser
- Repository Layer Separation: IConversationRepository separated from IDocumentRepository
- Service Layer Refactoring: AI, Database, Search, Parser, and Support services decomposed
- Model Consolidation: DatabaseSchema and DatabaseSchemaInfo unified

NEW FEATURES:
- SearchOptions: Per-request search configuration with flag-based filtering (-db, -d, -a, -i)
- FeatureToggles: Global feature flag configuration
- ContextExpansionService: Adjacent chunk context expansion
- FileParserResult: Standardized parser result structure
- DatabaseFileParser: SQLite database file parsing support
- Native Library Inclusion: Tesseract OCR native libraries bundled (Windows, macOS, Linux)
- Nullable Reference Types: Enhanced null safety across 14+ files
- ClearAllAsync Methods: Efficient bulk deletion operations
- Collection Management: Enhanced Qdrant collection operations
- Tesseract On-Demand Language Data: Automatic language support (30+ languages)
- Currency Symbol Correction: Improved OCR accuracy for financial documents
- Parallel Batch Processing: Ollama embeddings performance optimization
- Query Tokens Parameter: Pre-computed token support for performance
- Native Qdrant Text Search: Token-based filtering for improved search performance

IMPROVEMENTS:
- Language-Agnostic Code: Removed domain-specific and language-specific references
- Unicode Normalization: Better text retrieval across all languages
- PDF OCR Encoding: Automatic fallback handling
- Numbered List Detection: Improved counting query accuracy
- RAG Scoring: Enhanced relevance calculation with unique keyword bonus
- Document Search: Adaptive threshold and chunk prioritization
- Prompt Builder: Enhanced AI answer generation
- Qdrant Performance: Optimized GetAllAsync implementation

FIXES:
- Table Alias Enforcement: Prevents ambiguous column errors in SQL generation
- EnableDatabaseSearch Config: Proper feature flag handling
- macOS Native Libraries: OCR library inclusion and DYLD_LIBRARY_PATH configuration
- Missing Method Signature: DocumentSearchService restoration

CODE QUALITY:
- Maintained 0 errors, 0 warnings
- Full SOLID compliance
- Improved testability and extensibility
- Language-agnostic codebase

This release introduces significant architectural improvements and new features while maintaining full backward compatibility.