RecursiveAI.OneNlp.Models.Multilingual.Large 1.0.0

dotnet add package RecursiveAI.OneNlp.Models.Multilingual.Large --version 1.0.0
                    
NuGet\Install-Package RecursiveAI.OneNlp.Models.Multilingual.Large -Version 1.0.0
                    
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="RecursiveAI.OneNlp.Models.Multilingual.Large" Version="1.0.0" />
                    
For projects that support PackageReference, copy this XML node into the project file to reference the package.
<PackageVersion Include="RecursiveAI.OneNlp.Models.Multilingual.Large" Version="1.0.0" />
                    
Directory.Packages.props
<PackageReference Include="RecursiveAI.OneNlp.Models.Multilingual.Large" />
                    
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 RecursiveAI.OneNlp.Models.Multilingual.Large --version 1.0.0
                    
#r "nuget: RecursiveAI.OneNlp.Models.Multilingual.Large, 1.0.0"
                    
#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 RecursiveAI.OneNlp.Models.Multilingual.Large@1.0.0
                    
#: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=RecursiveAI.OneNlp.Models.Multilingual.Large&version=1.0.0
                    
Install as a Cake Addin
#tool nuget:?package=RecursiveAI.OneNlp.Models.Multilingual.Large&version=1.0.0
                    
Install as a Cake Tool

RecursiveAI.OneNlp.Models.Multilingual.Large

spaCy-class NLP for .NET. One fine-tuned mmBERT-base encoder with UD heads for 84 languages and entity heads for 19. Tags, morphology, lemmas, dependencies and sentence starts are supplied where the training treebanks have those annotations. Japanese, Classical Chinese, Thai, Vietnamese and Chinese include trained word segmentation. No downloads and no native dependencies. Unaffiliated with other projects named OneNLP.

Reference this package only. Its 18 component dependencies keep every model package under 250 MB without reducing any tensor: six byte-preserving pieces of the encoder graph, the Chinese heads, and 11 additional language-head groups. Installed weights total about 3.5 GB. The shared encoder loads on first use; a language's heads load only when requested. The additional head groups are package assets copied to weights/multilingual-large/heads/ in the build and publish directories; deploy the complete publish directory, not just the executable. The graph and original nine-language heads remain embedded. This layout supports NativeAOT without combining all weights into one resource image.

The plain Nlp verbs prefer a referenced fast-tier model for its language, otherwise this model. Explicit accurate-tier requests prefer this base-size package over the distilled multilingual package. The distilled package covers only the original nine languages; this package covers the 84 listed by MultilingualLargeModel.Languages. Language detection and UD coverage are separate: provide a language explicitly for text that the language detector does not cover.

using RecursiveAI.OneNlp;
using RecursiveAI.OneNlp.Neural.Components;

using var nlp = Pipeline.Create(Language.FromCode("cs"), Tier.Accurate)
    .Add<Morphologizer>().Add<Lemmatizer>().Add<DependencyParser>();
Doc doc = nlp.Process("Praha je hlavní město České republiky.");
foreach (Token token in doc)
{
    Console.WriteLine($"{token}: {token.Pos}, {token.Lemma}, {token.Dep}");
}

These are development weights. Published-reference quality gaps, per language and metric, are reported in the repository's docs/accurate-tier-rows.md, not hidden behind a multilingual average. Training data, licences, hashes, recipe settings and evaluation roles are in PROVENANCE.md, corpus.json, the included manifests and THIRD-PARTY-NOTICES.md.

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

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
1.0.0 0 11/8/2026

See CHANGELOG.md in the repository.