LM-Kit.NET
2026.8.5
Prefix Reserved
dotnet add package LM-Kit.NET --version 2026.8.5
NuGet\Install-Package LM-Kit.NET -Version 2026.8.5
<PackageReference Include="LM-Kit.NET" Version="2026.8.5" />
<PackageVersion Include="LM-Kit.NET" Version="2026.8.5" />
<PackageReference Include="LM-Kit.NET" />
paket add LM-Kit.NET --version 2026.8.5
#r "nuget: LM-Kit.NET, 2026.8.5"
#:package LM-Kit.NET@2026.8.5
#addin nuget:?package=LM-Kit.NET&version=2026.8.5
#tool nuget:?package=LM-Kit.NET&version=2026.8.5
LM-Kit.NET
Private AI for documents and internal knowledge, embedded in .NET.
LM-Kit.NET runs the full AI stack inside your process: agents, RAG, OCR, structured extraction, vision, speech and embeddings on curated open-weight models. No cloud calls, no Python sidecar, no per-token bill. Documents, indexes and prompts never leave the machine.
What you build with it
- Structured extraction: define fields, get schema-valid JSON from PDFs, scans and photographs, with per-field confidence.
- Cited answers: hybrid retrieval over your documents; answers carry document, page and score.
- Governed agents: planning, tool calling, orchestration and memory, with per-tool permissions and approval checkpoints.
- Document operations: classify, split, convert and permanently redact under human review.
- Vision, speech and embeddings: multimodal chat, local transcription, embedding models for search.
Extract fields from an invoice
using LMKit.Data;
using LMKit.Extraction;
using LMKit.Model;
var model = LM.LoadFromModelID("qwen3.5:4b");
var extractor = new TextExtraction(model);
extractor.Elements = new List<TextExtractionElement>
{
new("invoice_number", ElementType.String),
new("invoice_date", ElementType.Date),
new("total_amount", ElementType.Double),
};
extractor.SetContent(new Attachment("invoice.pdf"));
var data = extractor.Parse();
Console.WriteLine(data.Json);
Console.WriteLine(data.GetConfidence("total_amount"));
Put a governed agent to work
using LMKit.Agents;
using LMKit.Agents.Tools.BuiltIn;
using LMKit.Model;
var model = LM.LoadFromModelID("qwen3.5:9b");
var agent = Agent.CreateBuilder(model)
.WithTools(t =>
{
t.Register(BuiltInTools.WebSearch);
t.Register(BuiltInTools.CalcArithmetic);
})
.Build();
var result = await agent.RunAsync(
"Current BTC price in EUR, rounded to the nearest euro?");
Console.WriteLine(result.Text);
Runs on
- .NET Standard 2.0, .NET 8, 9 and 10
- Windows, Linux x64 and ARM64, macOS
- CPU out of the box; CUDA 12/13, Vulkan and Metal are used when present
Models download once and cache locally: chat, vision, OCR, embedding and speech families, plus your own fine-tunes.
The engine underneath
Dynamic Sampling, LM-Kit's adaptive inference layer, is why small local models hold their own on extraction, classification and structured generation: grammar-constrained output always parses, and it works on any supported model as shipped, including your fine-tunes.
Two ways to run the same engine
- Embed it with this package, in process, from C# or VB.NET.
- Deploy LM-Kit One, a private AI server with HTTP APIs and governed MCP tools, so AI assistants can use your document capabilities without reaching your files.
Links
Free for small teams, commercial licensing above that line: licensing.
| Product | Versions 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 is compatible. 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 is compatible. 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 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. |
| .NET Core | netcoreapp2.0 was computed. netcoreapp2.1 was computed. netcoreapp2.2 was computed. netcoreapp3.0 was computed. netcoreapp3.1 was computed. |
| .NET Standard | netstandard2.0 is compatible. netstandard2.1 was computed. |
| .NET Framework | net461 was computed. net462 was computed. net463 was computed. net47 was computed. net471 was computed. net472 was computed. net48 was computed. net481 was computed. |
| MonoAndroid | monoandroid was computed. |
| MonoMac | monomac was computed. |
| MonoTouch | monotouch was computed. |
| Tizen | tizen40 was computed. tizen60 was computed. |
| Xamarin.iOS | xamarinios was computed. |
| Xamarin.Mac | xamarinmac was computed. |
| Xamarin.TVOS | xamarintvos was computed. |
| Xamarin.WatchOS | xamarinwatchos was computed. |
-
.NETStandard 2.0
- AngleSharp (>= 1.7.1)
- DocumentFormat.OpenXml (>= 3.5.1)
- LM-Kit.NET.Backend.Cpu (= 2026.8.5)
- LM-Kit.NET.Backend.Vulkan (= 2026.8.5)
- Microsoft.Bcl.AsyncInterfaces (>= 10.0.11)
- Microsoft.Bcl.Memory (>= 10.0.11)
- Microsoft.Data.Sqlite (>= 10.0.11)
- MimeKit (>= 4.17.0)
- System.Buffers (>= 4.6.1)
- System.Diagnostics.DiagnosticSource (>= 10.0.11)
- System.IO.Packaging (>= 10.0.11)
- System.Linq.Async (>= 7.0.0)
- System.Memory (>= 4.6.3)
- System.Numerics.Tensors (>= 10.0.11)
- System.Numerics.Vectors (>= 4.6.1)
- System.Runtime.CompilerServices.Unsafe (>= 6.1.2)
- System.Text.Encodings.Web (>= 10.0.11)
- System.Text.Json (>= 10.0.11)
- System.Threading.Tasks.Extensions (>= 4.6.3)
-
net10.0
- AngleSharp (>= 1.7.1)
- DocumentFormat.OpenXml (>= 3.5.1)
- LM-Kit.NET.Backend.Cpu (= 2026.8.5)
- LM-Kit.NET.Backend.Vulkan (= 2026.8.5)
- Microsoft.Data.Sqlite (>= 10.0.11)
- MimeKit (>= 4.17.0)
- System.IO.Packaging (>= 10.0.11)
- System.Numerics.Tensors (>= 10.0.11)
-
net8.0
- AngleSharp (>= 1.7.1)
- DocumentFormat.OpenXml (>= 3.5.1)
- LM-Kit.NET.Backend.Cpu (= 2026.8.5)
- LM-Kit.NET.Backend.Vulkan (= 2026.8.5)
- Microsoft.Data.Sqlite (>= 10.0.11)
- MimeKit (>= 4.17.0)
- System.IO.Packaging (>= 10.0.11)
- System.Numerics.Tensors (>= 10.0.11)
-
net9.0
- AngleSharp (>= 1.7.1)
- DocumentFormat.OpenXml (>= 3.5.1)
- LM-Kit.NET.Backend.Cpu (= 2026.8.5)
- LM-Kit.NET.Backend.Vulkan (= 2026.8.5)
- Microsoft.Data.Sqlite (>= 10.0.11)
- MimeKit (>= 4.17.0)
- System.IO.Packaging (>= 10.0.11)
- System.Numerics.Tensors (>= 10.0.11)
NuGet packages (6)
Showing the top 5 NuGet packages that depend on LM-Kit.NET:
| Package | Downloads |
|---|---|
|
LM-Kit.NET.Data.Connectors.Qdrant
LM-Kit.NET.Data.Connectors.Qdrant acts as a seamless integration bridge between LM-Kit.NET and Qdrant vector databases. |
|
|
LM-Kit.NET.SemanticKernel
LM-Kit.NET.SemanticKernel acts as a seamless integration bridge between LM-Kit.NET and Microsoft Semantic Kernel. |
|
|
LM-Kit.NET.ExtensionsAI
LM-Kit.NET.ExtensionsAI provides Microsoft.Extensions.AI integration for LM-Kit.NET, implementing IChatClient and IEmbeddingGenerator for local LLM inference through the standard .NET AI abstraction layer. |
|
|
LM-Kit.NET.Data.Connectors.PgVector
LM-Kit.NET.Data.Connectors.PgVector acts as a seamless integration bridge between LM-Kit.NET and PostgreSQL databases using the pgvector extension. |
|
|
Peak.Abp.Silver.LMKit
Package Description |
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 2026.8.5 | 18 | 8/19/2026 |
| 2026.8.4 | 77 | 8/18/2026 |
| 2026.8.3 | 106 | 8/13/2026 |
| 2026.8.2 | 335 | 8/7/2026 |
| 2026.8.1 | 177 | 8/2/2026 |
| 2026.7.4 | 764 | 7/22/2026 |
| 2026.7.3 | 244 | 7/17/2026 |
| 2026.7.2 | 235 | 7/9/2026 |
| 2026.7.1 | 231 | 7/1/2026 |
| 2026.6.4 | 244 | 6/22/2026 |
| 2026.6.3 | 287 | 6/15/2026 |
| 2026.6.2 | 181 | 6/9/2026 |
| 2026.6.1 | 141 | 6/8/2026 |
| 2026.5.6 | 340 | 5/31/2026 |
| 2026.5.5 | 199 | 5/28/2026 |
| 2026.5.4 | 185 | 5/24/2026 |
| 2026.5.3 | 392 | 5/18/2026 |
| 2026.5.2 | 323 | 5/10/2026 |
| 2026.5.1 | 204 | 5/6/2026 |
| 2026.4.4 | 377 | 4/23/2026 |