Neural Speed

neural-speed is an innovative library developed by Intel to enhance the efficiency of Large Language Model (LLM) inference through low-bit quantization techniques. By reducing the precision of model weights and activations, neural-speed aims to accelerate inference while maintaining model accuracy, making it suitable for deployment in resource-constrained environments.