Dnn.LanguageModelConfig Module Reference

Classes
struct	Mila::Dnn::LanguageModelConfig< TDerived >
	CRTP base configuration for all deployable Mila language models. More...

Enumerations
enum class	Mila::Dnn::KvCacheCompression { None , FP8 }
	KV cache storage and compression strategy for GroupedQueryAttention. More...
enum class	Mila::Dnn::WeightQuantization { None , FP8 , FP4 }
	Weight storage and matmul strategy for Linear components. More...

Functions
std::string	Mila::Dnn::weightQuantizationName (WeightQuantization quantization)
	The scheme name recorded in an artifact and in its manifest.

Files
file	Mila/Src/Dnn/Core/LanguageModelConfig.ixx
	CRTP base configuration for all deployable Mila language models.