Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).

Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads

Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training

Code for the ICLR 2023 paper "GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers".

ChatGLM-6B: An Open Bilingual Dialogue Language Model | 开源双语对话语言模型

⚡LLM Zoo is a project that provides data, models, and evaluation benchmark for large language models.⚡

中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)

xk-time 是时间转换，时间计算，时间格式化，时间解析，日历，时间cron表达式和时间NLP等的工具，使用Java8（JSR-310），线程安全，简单易用，多达70几种常用日期格式化模板，支持Java8时间类和Date，轻量级，无第三方依赖。

[ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models

CPT: A Pre-Trained Unbalanced Transformer for Both Chinese Language Understanding and Generation

中文 NLP 预处理、解析工具包，准确、高效、易用 A Chinese NLP Preprocessing & Parsing Package www.jionlp.com

中文语句中的时间语义识别。即通过分析中文语句，识别出话语中提到的时间。

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

ONNX-TensorRT: TensorRT backend for ONNX

百度NLP：分词，词性标注，命名实体识别，词重要性

Transformer related optimization, including BERT, GPT

This repository contains the code for "Generating Datasets with Pretrained Language Models".

[EMNLP 2021] SimCSE: Simple Contrastive Learning of Sentence Embeddings https://arxiv.org/abs/2104.08821

Pretrained language model and its related optimization techniques developed by Huawei Noah's Ark Lab.

Implementation of RETRO, Deepmind's Retrieval based Attention net, in Pytorch

Google Research

Little python library for retrofitting autoregressive decoder transformers to use DeepMinds Retro framework: https://arxiv.org/pdf/2112.04426.pdf

A Keras TensorFlow 2.0 implementation of BERT, ALBERT and adapter-BERT.

Easy and Efficient Transformer : Scalable Inference Solution For Large NLP model

[ICML'21 Oral] I-BERT: Integer-only BERT Quantization

Block-sparse primitives for PyTorch

ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

Bolt is a deep learning library with high performance and heterogeneous flexibility.

Accelerated NLP pipelines for fast inference on CPU. Built with Transformers and ONNX runtime.