LMSupply.Reranker 0.24.0

.NET 10.0 This package targets .NET 10.0. The package is compatible with this framework or higher. 
dotnet add package LMSupply.Reranker --version 0.24.0
                    


                    

                
NuGet\Install-Package LMSupply.Reranker -Version 0.24.0
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="LMSupply.Reranker" Version="0.24.0" />
For projects that support PackageReference, copy this XML node into the project file to reference the package. 
<PackageVersion Include="LMSupply.Reranker" Version="0.24.0" />
                    


                            Directory.Packages.props
                        

                    

                
<PackageReference Include="LMSupply.Reranker" />
                    


                            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 LMSupply.Reranker --version 0.24.0
                    


                    

                
#r "nuget: LMSupply.Reranker, 0.24.0"
#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 LMSupply.Reranker@0.24.0
#: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=LMSupply.Reranker&version=0.24.0
                    


                            Install as a Cake Addin
                        

                    

                
#tool nuget:?package=LMSupply.Reranker&version=0.24.0
                    


                            Install as a Cake Tool

LMSupply.Reranker

Local semantic reranking for .NET with cross-encoder models.

Features

  • Zero-config: Models download automatically from HuggingFace
  • GPU Acceleration: CUDA, DirectML (Windows), CoreML (macOS)
  • Cross-platform: Windows, Linux, macOS
  • RAG Integration: Perfect for improving retrieval quality
  • Multi-Tokenizer Support: Automatic detection of WordPiece, Unigram, BPE tokenizers

Quick Start

using LMSupply.Reranker;

// Load the default model
await using var reranker = await LocalReranker.LoadAsync("default");

// Rerank documents
var results = await reranker.RerankAsync(
    query: "What is machine learning?",
    documents: ["ML is a branch of AI...", "The weather is nice..."],
    topK: 5);

foreach (var result in results)
    Console.WriteLine($"{result.Index}: {result.Score:F3}");

Available Models

Alias Model Size Tokenizer Description
default MS MARCO MiniLM L6 ~90MB WordPiece Best speed/quality balance
fast MS MARCO TinyBERT ~18MB WordPiece Ultra-fast, latency-critical
quality BGE Reranker Base ~440MB Unigram Higher accuracy, multilingual
large BGE Reranker Large ~1.1GB Unigram Highest accuracy
multilingual BGE Reranker v2-m3 ~1.1GB Unigram 8K context, 100+ languages

Tokenizer Auto-Detection

The reranker automatically detects the correct tokenizer type:

Type Detection Example Models
WordPiece vocab.txt MS MARCO MiniLM, TinyBERT
Unigram tokenizer.json (type: Unigram) bge-reranker-base, XLM-RoBERTa
BPE tokenizer.json (type: BPE) Some multilingual models

This ensures compatibility with virtually any cross-encoder model from HuggingFace.

GPU Acceleration

# NVIDIA GPU
dotnet add package Microsoft.ML.OnnxRuntime.Gpu

# Windows (AMD/Intel/NVIDIA)
dotnet add package Microsoft.ML.OnnxRuntime.DirectML

Configuration

var options = new RerankerOptions
{
    Provider = ExecutionProvider.Auto,  // GPU auto-detection
    MaxSequenceLength = 512,
    BatchSize = 32
};

var reranker = await LocalReranker.LoadAsync("default", options);

Version History

v0.8.9

  • Fixed JSON parsing error for Unigram vocab Array format (tuple-like arrays ["token", score])

v0.8.7

  • Added automatic tokenizer type detection (WordPiece, Unigram, BPE)
  • Fixed compatibility with bge-reranker-base and other Unigram-based models

v0.8.6

  • Fixed vocab parsing for Array vs Object format
Product Compatible and additional computed target framework versions.
.NET net10.0 is compatible.  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. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages (2)

Showing the top 2 NuGet packages that depend on LMSupply.Reranker:

Package Downloads
FluxIndex.Providers.LMSupply

LMSupply local AI embedding, reranking, and text completion provider for FluxIndex
IronHive.Cli.Core

IronHive CLI Core - Agent loop, tools, session management, and provider integrations for building AI-powered CLI tools

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
0.24.0 157 3/9/2026
0.22.2 93 3/9/2026
0.22.1 118 3/8/2026
0.22.0 82 3/8/2026
0.21.0 88 3/8/2026
0.20.2 83 3/7/2026
0.20.1 75 3/7/2026
0.20.0 84 3/7/2026
0.19.1 83 3/6/2026
0.19.0 78 3/6/2026
0.18.0 81 3/6/2026
0.17.2 80 3/5/2026
0.17.1 81 3/5/2026
0.17.0 80 3/5/2026
0.16.1 88 3/5/2026
0.16.0 79 3/5/2026
0.15.1 79 3/4/2026
0.15.0 87 3/3/2026
0.14.0 178 3/3/2026
0.13.9 121 2/27/2026
Loading failed