Sijun

hive. Python

OpenViking. Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

tinker-cookbook. Post-training with Tinker

terminal-bench-rl. GRPO training code which scales to 32xH100s for long horizon terminal/coding tasks. Base agent is now the top Qwen3 agent on Stanford's TerminalBench leaderboard.

NLWeb. Main reference implementation for NLWeb, implemented in Python.

dynamo. A Datacenter Scale Distributed Inference Serving Framework

claude-code. Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.

rllm. Democratizing Reinforcement Learning for LLMs

PAE. Python

Awesome-LVLM-Attack. 😎 up-to-date & curated list of awesome Attacks on Large-Vision-Language-Models papers, methods & resources.

PRIME. Scalable RL solution for advanced reasoning of language models

WebRL. Building Open LLM Web Agents with Self-Evolving Online Curriculum RL

RL4LMs. A modular RL library to fine-tune language models to human preferences

OpenRLHF. An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)

MathBlackBox. Python

JudgeBench. Python

swarm. Educational framework exploring ergonomic, lightweight multi-agent orchestration. Managed by OpenAI Solution team.

Awesome-Jailbreak-on-LLMs. Awesome-Jailbreak-on-LLMs is a collection of state-of-the-art, novel, exciting jailbreak methods on LLMs. It contains papers, codes, datasets, evaluations, and analyses.

g1. g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

ell. A language model programming library.

Awesome-LLM-Strawberry. A collection of LLM papers, blogs, and projects, with a focus on OpenAI o1 🍓 and reasoning techniques.

llama3. The official Meta Llama 3 GitHub site

lloco. The official repo for "LLoCo: Learning Long Contexts Offline"

AutoCompressors. [EMNLP 2023] Adapting Language Models to Compress Long Contexts

llm-course. Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

search_with_lepton. Building a quick conversation-based search demo with Lepton AI.

Awesome-LLM-Safety. A curated list of safety-related papers, articles, and resources focused on Large Language Models (LLMs). This repository aims to provide researchers, practitioners, and enthusiasts with insights into the safety implications, challenges, and advancements surrounding these powerful models.

AutoGPT. AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.

rust-course. 什么？你敢放心的把后背交给 AI? 我赌你不敢，那就来学学 AI 时代最酷、最安全、最快的语言吧。本书拥有全面且深入的讲解、生动贴切的示例、德芙般丝滑的内容，这可能是目前最用心的 Rust 中文学习教程 / Book

simplepir. The reference implementation of SimplePIR and DoublePIR, accompanying the paper "One Server for the Price of Two: Simple and Fast Single-Server Private Information Retrieval" (USENIX Security 2023).

fbpcs. FBPCS (Facebook Private Computation Solutions) leverages secure multi-party computation (MPC) to output aggregated data without making unencrypted, readable data available to the other party or any third parties. Facebook provides impression & opportunity data, and the advertiser provides conversion / outcome data.

fbpcp. FBPCP (Facebook Private Computation Platform) is a secure, privacy safe and scalable architecture to deploy MPC (Multi Party Computation) applications in a distributed way on virtual private clouds. FBPCF (Facebook Private Computation Framework) is for scaling MPC computation up via threading, while FBPCP is for scaling MPC computation out via Private Scaling architecture.

Private-ID. A collection of algorithms that can do join between two parties while preserving the privacy of keys on which the join happens

fbpcf. Private computation framework library allows developers to perform randomized controlled trials, without leaking information about who participated or what action an individual took. It uses secure multiparty computation to guarantee this privacy. It is suitable for conducting A/B testing, or measuring advertising lift and learning the aggregate statistics without sharing information on the individual level.

LSC. Least Square Calibration for Peer Reviews

bellman. zk-SNARK library.

PokerRL. Framework for Multi-Agent Deep Reinforcement Learning in Poker

csprng. Cryptographically secure pseudorandom number generators for PyTorch

rebel. An algorithm that generalizes the paradigm of self-play reinforcement learning and search to imperfect-information games.

mina. Mina is a cryptocurrency protocol with a constant size blockchain, improving scaling while maintaining decentralization and security.

Dance_Dance_Generation. Pytorch implementation of Dance Dance Generation: Motion Transfer for Internet Videos

jukebox. Code for the paper "Jukebox: A Generative Model for Music"

libiop. C++ library for IOP-based zkSNARKs

poly-commit. A Rust library for polynomial commitments

libsnark. C++ library for zkSNARKs

bitcoin. Bitcoin Core integration/staging tree

fairseq. Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

CrypTen. A framework for Privacy Preserving Machine Learning

awesome-game-ai. Awesome Game AI materials of Multi-Agent Reinforcement Learning

detectron2. Detectron2 is a platform for object detection, segmentation and other visual recognition tasks.

BERT-pytorch. Google AI 2018 BERT pytorch implementation

flair. A very simple framework for state-of-the-art Natural Language Processing (NLP)

deeplab-pytorch. PyTorch re-implementation of DeepLab v2 on COCO-Stuff / PASCAL VOC datasets

deep-photo-styletransfer. Code and data for paper "Deep Photo Style Transfer": https://arxiv.org/abs/1703.07511

deep-painterly-harmonization. Code and data for paper "Deep Painterly Harmonization": https://arxiv.org/abs/1804.03189

detexify. Latex Symbol Classifier Web Frontend

detexify-hs-backend. Detexify Backend written in Haskell

thefuck. Magnificent app which corrects your previous console command.

FastMaskRCNN. Mask RCNN in TensorFlow

dl-docker. An all-in-one Docker image for deep learning. Contains all the popular DL frameworks (TensorFlow, Theano, Torch, Caffe, etc.)

fast-wavenet. Speedy Wavenet generation using dynamic programming :zap: