AI-Efficiency/IR-QLoRA

Statistically-Lossless Quantization of Large Language Models

Official implementation of LittleBit (NeurIPS 2025) and its follow-up LittleBit-2 (ICML 2026)

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

Official Pytorch repository for Extreme Compression of Large Language Models via Additive Quantization https://arxiv.org/pdf/2401.06118.pdf and PV-Tuning: Beyond Straight-Through Estimation for Extreme LLM Compression https://arxiv.org/abs/2405.14852

Gumbel-Softmax post-training quantization for LLMs (1–3 bit scalar, INT/GGUF-compatible).

An Efficient Compression Framework for LLM

Accessible large language models via k-bit quantization for PyTorch.

LLM quantization optimizer

AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation: