Jittor is a high-performance deep learning framework based on JIT compiling and meta-operators.

Sparsity-aware deep learning inference runtime for CPUs

torch-optimizer -- collection of optimizers for Pytorch

A Repo For Document AI

[NeurIPS'24] HippoRAG is a novel RAG framework inspired by human long-term memory that enables LLMs to continuously integrate knowledge across external documents. RAG + Knowledge Graphs + Personali…

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

LLaMA: Open and Efficient Foundation Language Models

Code and documents of LongLoRA and LongAlpaca (ICLR 2024 Oral)

A Deep Learning Recommender System

compiler learning resources collect.

HuixiangDou: Overcoming Group Chat Scenarios with LLM-based Technical Assistance

PPO x Family DRL Tutorial Course（决策智能入门级公开课：8节课帮你盘清算法理论，理顺代码逻辑，玩转决策AI应用实践 ）

百亿参数的中英文双语基座大模型

A deep matching model library for recommendations & advertising. It's easy to train models and to export representation vectors which can be used for ANN search.

xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism

Run Mixtral-8x7B models in Colab or consumer desktops

mPLUG-DocOwl: Modularized Multimodal Large Language Model for Document Understanding

GraphRAG using Local LLMs - Features robust API and multiple apps for Indexing/Prompt Tuning/Query/Chat/Visualizing/Etc. This is meant to be the ultimate GraphRAG/KG local LLM app.

TigerBot: A multi-language multi-task LLM

This includes the original implementation of SELF-RAG: Learning to Retrieve, Generate and Critique through self-reflection by Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi.

Schedule-Free Optimization in PyTorch

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

Awesome Deep Learning papers for industrial Search, Recommendation and Advertisement. They focus on Embedding, Matching, Pre-Ranking, Ranking (CTR/CVR prediction), Post Ranking, Relevance, LLM, Rei…

MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.

VideoSys: An easy and efficient system for video generation

推荐/广告/搜索领域工业界经典以及最前沿论文集合。A collection of industry classics and cutting-edge papers in the field of recommendation/advertising/search.

Pyserini is a Python toolkit for reproducible information retrieval research with sparse and dense representations.

Recommender Learning with Tensorflow2.x

MoBA: Mixture of Block Attention for Long-Context LLMs

[ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment