You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
intel/neural-compressor’s past year of commit activity
SOTA low-bit LLM quantization (INT8/FP8/MXFP8/INT4/MXFP4/NVFP4) & sparsity; leading model compression techniques on PyTorch, TensorFlow, and ONNX Runtime
A SOTA quantization algorithm for high-accuracy low-bit LLM inference, seamlessly optimized for CPU/XPU/CUDA, with multi-datatype support and full compatibility…
This project contains a set of plugins for PyTorch* ecosystem libraries which enable hardware acceleration on Intel® GPUs thru the XPU PyTorch* device backend.
An advanced tool leveraging LLM and RAG technology to empower faculty with curriculum building, teaching support, and personalized learning experiences.
Experimental reference implementation code for [BBB-]DDD-AES ciphers, accompanying the "Efficient Instances of Docked Double Decker with AES, and Application to…