export

Modules

modelopt.torch.export.distribute	torch.distribute utils.
modelopt.torch.export.hf_config_map	Define the config mapping between HF and modelopt.
modelopt.torch.export.layer_utils	Utils for model_config export.
modelopt.torch.export.mcore_config_map	Define the config mapping between Mcore and modelopt.
modelopt.torch.export.model_config	This module defines the model_config format.
modelopt.torch.export.model_config_export	Code that export optimized models to the TensorRT-LLM checkpoint.
modelopt.torch.export.model_config_utils	Common utils for the ModelConfig.
modelopt.torch.export.postprocess	Utils to load and process model_config.
modelopt.torch.export.quantization_utils	Utils for quantization including scaling factors adjustments.
modelopt.torch.export.tensorrt_llm_type	Code from TRT-LLM that export optimized models to the TensorRT-LLM checkpoint.
modelopt.torch.export.tensorrt_llm_utils	Utils for TensorRT-LLM checkpoint export.
modelopt.torch.export.transformer_engine	Convert the Model Optimizer quantized model to the transformer_engine.
modelopt.torch.export.unified_export_hf	Code that export quantized Hugging Face models for deployment.

Export package. So far it only supports selected nemo and huggingface LLMs.