AI-SOVEREIGN
Home / Open-source AI / neural-compressor
Open-source AI

neural-compressor

SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime

GitHub stars2.7K
Forks337
7d change–
LanguagePython
LicenseApache-2.0
Last commit20 Nov 2026
auto-tuningawqfp4gptqint4int8knowledge-distillationlarge-language-models

More from intel