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.
 
 
 
 
 
 
Martin Evans a0335f67a4 - Added `LLamaTemplate` which efficiently formats a series of messages according to the model template. 1 year ago
..
DecodeResult.cs Swapped `StatelessExecutor` to use `llama_decode`! 1 year ago
GGMLType.cs Added a comment on the type itself 1 year ago
GPUSplitMode.cs Updated everything to work with llama.cpp ce32060198b7e2d6a13a9b8e1e1369e3c295ae2a 1 year ago
GroupDisposable.cs - Added `GroupDisposable` to dispose a collection of items all together 1 year ago
LLamaBatch.cs Fixed off by one error in LLamaBatch sampling position (#626) 1 year ago
LLamaBeamView.cs April 2024 Binary Update (#662) 1 year ago
LLamaBeamsState.cs April 2024 Binary Update (#662) 1 year ago
LLamaChatMessage.cs - Added `LLamaTemplate` which efficiently formats a series of messages according to the model template. 1 year ago
LLamaContextParams.cs - Added `LLamaWeights.LoadFromFileAsync`. 1 year ago
LLamaFtype.cs April 2024 Binary Update (#662) 1 year ago
LLamaGrammarElement.cs Improved test coverage. Discovered some issues: 2 years ago
LLamaKvCacheView.cs April 2024 Binary Update (#662) 1 year ago
LLamaLogLevel.cs Logging interceptor (#649) 1 year ago
LLamaModelMetadataOverride.cs March Binary Update (#565) 1 year ago
LLamaModelParams.cs - Added `LLamaWeights.LoadFromFileAsync`. 1 year ago
LLamaModelQuantizeParams.cs April 2024 Binary Update (#662) 1 year ago
LLamaNativeBatch.cs Removed `LLamaBatchSafeHandle` (using unmanaged memory, created by llama.cpp) and replaced it with a fully managed `LLamaBatch`. Modified the `BatchedDecoding` example to use new managed batch. 1 year ago
LLamaPoolingType.cs April 2024 Binary Update (#662) 1 year ago
LLamaPos.cs Added increment and decrement operators to `LLamaPos` 1 year ago
LLamaRopeType.cs March Binary Update (#565) 1 year ago
LLamaSeqId.cs Removed `LLamaBatchSafeHandle` (using unmanaged memory, created by llama.cpp) and replaced it with a fully managed `LLamaBatch`. Modified the `BatchedDecoding` example to use new managed batch. 1 year ago
LLamaToken.cs Added a test with examples of troublesome strings from 0.9.1 1 year ago
LLamaTokenData.cs made casts to/from int explicit, fixed places affected 1 year ago
LLamaTokenDataArray.cs Classifier Free Guidance (#536) 1 year ago
LLamaTokenType.cs Assorted small changes to clean up some code warnings 1 year ago
LLamaVocabType.cs April 2024 Binary Update (#662) 1 year ago
LLavaImageEmbed.cs Llava api (#563) 1 year ago
NativeApi.BeamSearch.cs Code cleanup driven by R# suggestions: 1 year ago
NativeApi.Grammar.cs - Removed unnecessary constructors from safe handles 1 year ago
NativeApi.LLava.cs fix: typos. 1 year ago
NativeApi.Load.cs fix: typos. 1 year ago
NativeApi.Quantize.cs Modified `llama_model_quantize` to accept argument by `ref` instead of pointer. 1 year ago
NativeApi.Sampling.cs Classifier Free Guidance (#536) 1 year ago
NativeApi.cs - Added `LLamaTemplate` which efficiently formats a series of messages according to the model template. 1 year ago
NativeLibraryConfig.cs fix: typos. 1 year ago
NativeLogConfig.cs Logging interceptor (#649) 1 year ago
RopeScalingType.cs March Binary Update (#565) 1 year ago
SafeLLamaContextHandle.cs Added checks in `Decode` to skip doing anything if the batch is empty. 1 year ago
SafeLLamaGrammarHandle.cs Using `is` check instead of `== null` 1 year ago
SafeLLamaHandleBase.cs - Removed one of the constructors of `SafeLLamaHandleBase`, which implicitly states that memory is owned. Better to be explicit about this kind of thing! 1 year ago
SafeLlamaModelHandle.cs - Added `LLamaWeights.LoadFromFileAsync`. 1 year ago
SafeLlavaImageEmbedHandle.cs - Removed unnecessary constructors from safe handles 1 year ago
SafeLlavaModelHandle.cs - Added `LoadFromFileAsync` method for `LLavaWeights` 1 year ago