qtensor

Modules

modelopt.torch.quantization.qtensor.base_qtensor

Base Class for Real Quantized Tensor.

modelopt.torch.quantization.qtensor.int4_tensor

Implements INT4 quantization for efficient tensor storage and computation.

modelopt.torch.quantization.qtensor.nf4_tensor

Implements NF4 quantization for efficient tensor storage and computation.

modelopt.torch.quantization.qtensor.nvfp4_tensor

Implements NVFP4 quantization for efficient tensor storage and computation.

Tensor Class for Real Quantization.