export

Modules

modelopt.torch.export.convert_hf_config	Convert modelopt quantization export config to align with llm-compressor config format.
modelopt.torch.export.diffusers_utils	Code that export quantized Hugging Face models for deployment.
modelopt.torch.export.hf_export_handlers	Built-in module handlers for unified Hugging Face export.
modelopt.torch.export.layer_utils	Module-shape predicates and MoE quantizer helpers shared by every export backend.
modelopt.torch.export.layerwise_export	Write each decoder layer's quantized checkpoint shard as soon as it is calibrated.
modelopt.torch.export.model_utils	Utility functions for model type detection and classification.
modelopt.torch.export.moe_utils	Utilities for Mixture-of-Experts (MoE) model export.
modelopt.torch.export.plugins	Export package plugin.
modelopt.torch.export.quant_aware_conversion	Quantization-aware reverse weight conversion for unified HF export.
modelopt.torch.export.quant_format	The quantization and KV-cache format names shared by every export backend.
modelopt.torch.export.quant_utils	Utils for quantization including scaling factors adjustments.
modelopt.torch.export.registry	Registries dispatching per-module logic for the unified HF export path.
modelopt.torch.export.shard_cast_utils	Shared primitives for streaming checkpoint casts and export.
modelopt.torch.export.transformer_engine	Convert the Model Optimizer quantized model to the transformer_engine.
modelopt.torch.export.trtllm	Checkpoint export logic for the TensorRT-LLM specific format.
modelopt.torch.export.unified_export_hf	Code that export quantized Hugging Face models for deployment.
modelopt.torch.export.unified_export_hf_streaming	Streaming HF checkpoint export for disk/CPU-offloaded models.
modelopt.torch.export.unified_export_megatron	Code that export quantized Megatron Core models for deployment.

Export package for Hugging Face and Megatron-based models.