net10.0
Shared LLamaSharp/llama.cpp runtime management for LMSupply. Provides on-demand native binary downloading, GPU detection, and fallback chain support.
Published
Jan 19, 2026
9mo ago
Total Downloads
28.1K
28,053
Install
dotnet add package LMSupply.Llama --version 0.9.2
| Product | Versions |
|---|---|
| .NET |
README
LMSupply.Llama
Shared LLamaSharp/llama.cpp runtime management for LMSupply libraries.
Features
- On-demand binary downloading: Native binaries are downloaded from NuGet on first use
- GPU auto-detection: Automatic selection of CUDA, Vulkan, Metal, or CPU backend
- Fallback chain: CUDA 13 → CUDA 12 → Vulkan → CPU (platform-dependent)
- Caching: Downloaded binaries are cached locally for reuse
Supported Backends
| Backend | Platform | GPU |
|---|---|---|
| CUDA 12/13 | Windows, Linux | NVIDIA |
| Vulkan | Windows, Linux | AMD, Intel |
| Metal | macOS | Apple Silicon |
| CPU | All | AVX2/AVX512 optimized |
Usage
This package is used internally by:
LMSupply.Generator- GGUF text generationLMSupply.Embedder- GGUF embedding
Architecture
LMSupply.Llama
├── LlamaBackend.cs # Backend enum (Cpu, Cuda12, Vulkan, Metal, etc.)
├── LlamaRuntimeManager.cs # Singleton manager with fallback chain
├── LlamaNuGetDownloader.cs # Downloads LLamaSharp.Backend.* packages
└── DownloadProgress.cs # Progress reporting
Changes from Previous Version
No structural changes between v0.9.1 and v0.9.2.
Adoption Assessment
Assess this package against its supported frameworks, license, maintenance signals and dependency graph before making it a project standard.
Compatibility metadata lists 1 target framework; confirm the exact target used by your application.
Related Guides & Workflows
Learn how to evaluate, integrate, or audit this package in your .NET project.