OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Local & Self-hosted AI

Llamasharp

LLamaSharp is a C#/.NET library based on llama.cpp that lets developers run LLaMA-family and LLaVA language models locally in their applications. It uses native CPU or GPU backends and accepts GGUF model files.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Llamasharp — OpenViST project profile in Local & Self-hosted AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use LLamaSharp when a .NET application needs to run a supported model locally and you can provide a compatible GGUF file and native backend.

Who it is for

  • C# and .NET developers adding local language-model inference to an application.
  • Teams building local chatbots or other applications around supported LLaMA models.
  • Developers exploring local LLaVA inference in a C# application.

Documented uses

  • Add local language-model capabilities to a C# application.
  • Build a local chatbot using a supported LLaMA model.
  • Use a native CPU or GPU backend for inference within a .NET project.
Before you choose

Requirements & limits

The repository identifies LLamaSharp as MIT-licensed and publishes installation through NuGet.

  • LLamaSharp requires GGUF model files; other model formats need conversion.
  • A matching native backend is required, and available package options depend on platform and hardware.
  • The README lists CPU support for Windows, Linux and Mac; CUDA 11/12 and Vulkan packages for Windows and Linux; and Metal GPU support through the CPU package on Mac.
  • The repository describes higher-level APIs and RAG support, while integrations such as LangChain are maintained in separate repositories.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install LLamaSharp with a matching native backend

Prepare a C#/.NET project with LLamaSharp and a backend package for local model use.

Before you start

  • A C#/.NET project that can install NuGet packages.
  • A target platform and backend choice supported by the listed packages.

Steps

  1. Install the LLamaSharp NuGet package using Install-Package LLamaSharp.

  2. Install a backend package suited to the target platform: CPU supports Windows, Linux, and Mac; CUDA 11 or 12 and Vulkan support Windows and Linux; the CPU package also provides Metal GPU support on Mac.

Expected outcome

The project has LLamaSharp and a selected native backend package installed, ready for model preparation.

Guide limitations

  • Backend packages use native libraries; select a backend compatible with both the device and LLamaSharp version.
  • If no published backend matches the device, the project describes compiling a backend as an alternative.
  • This setup does not configure or execute a model.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
7 October 2026
Original website title
SciSharp/LLamaSharp · Official homepage title · retained in its source language
How we review sourcesSuggest a correction