Skip to main content

Module models

Module models 

Source
Expand description

Model-related types and configurations

Structs§

AttentionConfig
Attention configuration for model architecture
ModelConfig
Model configuration for runtime
ModelInfo
Model information and metadata
ModelMemoryRequirements
Memory requirements for model inference
RopeScaling
RoPE (Rotary Position Embedding) scaling configuration
TokenUsage
Token usage statistics

Enums§

Activation
Activation function type
ModelSource
Model loading source specification
ModelType
Model type enumeration
NormType
Normalization type used in the model
QuantizationConfig
Quantization configuration