You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 
Yaohui Liu bcd4c5605b
feat: add n_gpu_layers and prompt_cache_all params.
3 years ago
..
Exceptions feat: add the framework and basic usages. 3 years ago
Native refactor: use official api of quantization instead. 3 years ago
runtimes build: optimize the building of LLama. 3 years ago
ChatSession.cs feat: enable history for chat session. 3 years ago
IChatModel.cs feat: add ChatSession. 3 years ago
LLamaCache.cs feat: add ChatSession. 3 years ago
LLamaEmbedder.cs fix: add IDisposable to model classes. 3 years ago
LLamaModel.cs feat: add n_gpu_layers and prompt_cache_all params. 3 years ago
LLamaModelV1.cs refactor: change some file names. 3 years ago
LLamaParams.cs feat: add n_gpu_layers and prompt_cache_all params. 3 years ago
LLamaSharp.csproj build: optimize the building of LLama. 3 years ago
LLamaState.cs feat: add the framework and basic usages. 3 years ago
LLamaTypes.cs feat: add ChatSession. 3 years ago
Logger.cs feat: add the framework and basic usages. 3 years ago
Quantizer.cs refactor: use official api of quantization instead. 3 years ago
Utils.cs feat: add ChatSession. 3 years ago
libllama.dll build: revise the building of master branch. 3 years ago
libllama.so build: revise the building of master branch. 3 years ago

C#/.NET上易用的LLM高性能推理框架,支持LLaMA和LLaVA系列模型。

C# Text Metal JavaScript HTML+Razor other