Liqun Chen

clawterminal-docs. OpenClaw Terminal — Guides & Tutorials

hermes-agent. The agent that grows with you

skillforge. Quality layer for Agent Skills (agentskills.io) — lint, hash, sign, version, eval your SKILL.md files. 10 structural checks, Ed25519 signing, semantic diff, 3-stage eval pipeline. Bidirectional SKILL.md ↔ AIF conversion.

claw-code-parity. Join Discord: https://discord.gg/5TUQKqFWd / claw-code Rust port parity work - it is temporary work while claw-code repo is doing migration

claw-code. An agent-managed museum exhibit, built in Rust with Gajae-Code / LazyCodex — developed and maintained with no human intervention.

markit. 🖍️ Convert anything to markdown. Mark it.

fast-weight-attention. Implementation of Fast Weight Attention

RIM-pytorch. Implementation of Recurrent Independent Mechanisms in Pytorch

skills_public. My Public Skills for Agents

CLI-Anything. "CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: https://clianything.cc/

pi-multi-pass. Multi-subscription extension for pi -- use multiple OAuth accounts per provider (Anthropic, Codex, Copilot, Gemini, Antigravity)

OpenClaw-RL. OpenClaw-RL: Train any agent simply by talking

llm_training_handbook. An open collection of methodologies to help with successful training of large language models.

autoresearch. AI agents running research on single-GPU nanochat training automatically

superpowers. An agentic skills framework & software development methodology that works.

LLMs-from-scratch. Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Megatron-LM. Ongoing research training transformer models at scale

openclaw. Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

awesome-claude-skills. A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

tiny8. A tiny CPU simulator written in Python

gepa. Optimize prompts, code, and more with AI-powered Reflective Optimization

tokenizer-ui. HTML

spacy-models. 💫 Models for the spaCy Natural Language Processing (NLP) library

AI-Trader. "AI-Trader: 100% Fully-Automated Agent-Native Trading"

openfold-3. A fully open source biomolecular structure prediction model based on AlphaFold3

nof0. NOF0 - 开源的 AI 交易竞技场

DeepSeek-OCR. Contexts Optical Compression

nanochat. The best ChatGPT that $100 can buy.

dexter. An autonomous agent for deep financial research

tinker-cookbook. Post-training with Tinker

Kronos. Kronos: A Foundation Model for the Language of Financial Markets

npq-vit. [ICLR 2025] Binary Spherical Quantization + [CVPR 2026] Leech Spherical Quantization

lookahead-keys-attention. Causal Attention with Lookahead Keys

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

DeepRetrieval. [COLM’25] DeepRetrieval — 🔥 Training Search Agent by RLVR with Retrieval Outcome

Qwen3-Coder. Qwen3-Coder is the code version of Qwen3, the large language model series developed by Qwen team.

smolagents. 🤗 smolagents: a barebones library for agents that think in code.

ASI-Arch. AlphaGo Moment for Model Architecture Discovery.

simple_GRPO. A very simple GRPO implement for reproducing r1-like LLM thinking.

ARMA-Attention. Code implementation of the paper "WAVE: Weighted Autoregressive Varying Gate for Time Series Forecasting" (ICML 2025)

gpt-oss. gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

CognitiveKernel-Pro. Deep Research Agent CognitiveKernel-Pro from Tencent AI Lab. Paper: https://arxiv.org/pdf/2508.00414

OLMoE. OLMoE: Open Mixture-of-Experts Language Models

nautilus_trader. Production-grade Rust-native trading engine with deterministic event-driven architecture

Wan2.2. Wan: Open and Advanced Large-Scale Video Generative Models

TimeBase. Python

HRM. Exploration into the proposed architecture from Sapient Intelligence of Singapore 🇸🇬

HRM. Hierarchical Reasoning Model Official Release

PruLong. Code for the preprint "Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?"

SAITS. The official PyTorch implementation of the paper "SAITS: Self-Attention-based Imputation for Time Series". A fast and state-of-the-art (SOTA) deep-learning neural network model for efficient time-series imputation (impute multivariate incomplete time series containing NaN missing data/values with machine learning). https://arxiv.org/abs/2202.08516

hnet. H-Net: Hierarchical Network with Dynamic Chunking

esm. Evolutionary Scale Modeling (esm): Pretrained language models for proteins

openai-imo-2025-proofs.

x-transformers-rl. Implementation of a transformer for reinforcement learning using `x-transformers`

medical-imaging-datasets. A list of Medical imaging datasets.

nanoVLM. The simplest, fastest repository for training/finetuning small-sized VLMs.

minirocket. MINIROCKET: A Very Fast (Almost) Deterministic Transform for Time Series Classification

ChinaTextbook. 所有小初高、大学PDF教材。

UCR_Time_Series_Classification_Deep_Learning_Baseline. Fully Convlutional Neural Networks for state-of-the-art time series classification

TabPFN. ⚡ TabPFN: Foundation Model for Tabular Data ⚡

FramePack. Lets make video diffusion practical!

Convolutional-KANs. This project extends the idea of the innovative architecture of Kolmogorov-Arnold Networks (KAN) to the Convolutional Layers, changing the classic linear transformation of the convolution to learnable non linear activations in each pixel.

TimeXer. Official implementation for "TimeXer: Empowering Transformers for Time Series Forecasting with Exogenous Variables" (NeurIPS 2024)

OpenManus. No fortress, purely open ground. OpenManus is Coming.

TimePFN. Official repository for "TimePFN: Effective Multivariate Time Series Forecasting with Synthetic Data" (AAAI 2025).

Q-Flow. Complete Reinforcement Learning Toolkit for Large Language Models!

KANbeFair. A More Fair and Comprehensive Comparison between KAN and MLP

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

KAN-Tutorial. Understanding Kolmogorov-Arnold Networks: A Tutorial Series on KAN using Toy Examples

deep-cross-attention. Implementation of the proposed DeepCrossAttention by Heddes et al at Google research, in Pytorch

convkan. Convolutional layer for Kolmogorov-Arnold Network (KAN)

efficient-kan. An efficient pure-PyTorch implementation of Kolmogorov-Arnold Network (KAN).

agents-course. This repository contains the Hugging Face Agents Course.

rational_kat_cu. Python

e2-tts-pytorch. Implementation of E2-TTS, "Embarrassingly Easy Fully Non-Autoregressive Zero-Shot TTS", in Pytorch

rational_activations. Rational Activation Functions - Replacing Padé Activation Units

kat. [ICLR2025] Kolmogorov-Arnold Transformer

flash-attention. Fast and memory-efficient exact attention

Neural-Experts. This repo contains the official code release of the Neural Experts paper, published in NeurIPS 2024.

open-thoughts. Fully open data curation for reasoning models

Janus. Janus-Series: Unified Multimodal Understanding and Generation Models

TinyZero. Minimal reproduction of DeepSeek R1-Zero

DeepSeek-R1.

micro_diffusion. Official repository for our work on micro-budget training of large-scale diffusion models.

DeepSeek-V3. Python

PRIME. Scalable RL solution for advanced reasoning of language models

coconut-pytorch. Implementation of 🥥 Coconut, Chain of Continuous Thought, in Pytorch

DiT. Official PyTorch Implementation of "Scalable Diffusion Models with Transformers"

large_concept_model. Large Concept Models: Language modeling in a sentence representation space

LVMAE-pytorch. Implementation of the proposed LVMAE, from the paper, Extending Video Masked Autoencoders to 128 frames, in Pytorch

Protenix. Toward High-Accuracy Open-Source Biomolecular Structure Prediction.

llm2vec. Code for 'LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders'

O1-Journey. O1 Replication Journey

deep-vector-quantization. VQVAEs, GumbelSoftmaxes and friends

LLM101n. LLM101n: Let's build a Storyteller

USACO. Can Language Models Solve Olympiad Programming?

TimeSiam. Python

samformer. Official implementation of SAMformer, a transformer leveraging Sharpness-Aware Minimization and Channel-Wise Attention for Time Series Forecasting.

rectified-flow-pytorch. Implementation of rectified flow and some of its followup research / improvements in Pytorch

PDF-Extract-Kit. A Comprehensive Toolkit for High-Quality PDF Content Extraction

PEER-pytorch. Pytorch implementation of the PEER block from the paper, Mixture of A Million Experts, by Xu Owen He at Deepmind

SparseTSF. [TPAMI 2026 & ICML 2024 Oral] Official repository of the SparseTSF paper: "SparseTSF: Modeling Long-term Time Series Forecasting with 1k Parameters". This work is developed by the Lab of Professor Weiwei Lin (linww@scut.edu.cn), South China University of Technology; Pengcheng Laboratory.

pg-is-all-you-need. Policy Gradient is all you need! A step-by-step tutorial for well-known PG methods.

KeyBERT. Minimal keyword extraction with BERT

build-nanogpt. Video+code lecture on building nanoGPT from scratch

sumy. Module for automatic summarization of text documents and HTML pages.

granite-tsfm. Foundation Models for Time Series

c-style. My favorite C programming practices.

qlib. Qlib is an AI-oriented Quant investment platform that aims to use AI tech to empower Quant Research, from exploring ideas to implementing productions. Qlib supports diverse ML modeling paradigms, including supervised learning, market dynamics modeling, and RL, and is now equipped with https://github.com/microsoft/RD-Agent to automate R&D process.

Time-Series-Forecasting-and-Deep-Learning. Resources about time series forecasting and deep learning.

ToRA. ToRA is a series of Tool-integrated Reasoning LLM Agents designed to solve challenging mathematical reasoning problems by interacting with tools [ICLR'24].

MAE-LM. [ICLR 2024] Representation Deficiency in Masked Language Modeling

alphafold3-pytorch. Implementation of Alphafold 3 from Google Deepmind in Pytorch

LLaVA-pp. 🔥🔥 LLaVA++: Extending LLaVA with Phi-3 and LLaMA-3 (LLaVA LLaMA-3, LLaVA Phi-3)

chronos-forecasting. Chronos: Pretrained Models for Time Series Forecasting

LLaVA-HR. [ICLR2025] LLaVA-HR: High-Resolution Large Language-Vision Assistant

selfcodealign. [NeurIPS'24] SelfCodeAlign: Self-Alignment for Code Generation

mistral-common. Official inference library for pre-processing of Mistral models

MGM. Official repo for "Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models"

LlamaFactory. Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

twenty. The open alternative to Salesforce, designed for AI.

llm.c. LLM training in simple, raw C/CUDA

Open-Sora-Plan. This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.

guanaco-lora. Instruct-tune LLaMA on consumer hardware

SWE-agent. SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

pytorch-vqvae. Vector Quantized VAEs - PyTorch Implementation

awesome-mixture-of-experts. A collection of AWESOME things about mixture-of-experts

ControlNet. Let us control diffusion models!

gpt-llm-trainer. Jupyter Notebook

mistral-inference. Official inference library for Mistral models

hackathon. Python

OpenHands. 🙌 OpenHands: AI-Driven Development

Open-Sora. Open-Sora: Democratizing Efficient Video Production for All

grok-1. Grok open release

launchables. Collection of notebook guides created by the Brev.dev team!

annotated-mamba. Annotated version of the Mamba paper

LOBCAST. LOBCAST is a Python-based open-source framework for stock market trend forecasting using Limit Order Book (LOB) data. 🤖📈

ml-sigma-reparam. Python

mixture-of-experts. A Pytorch implementation of Sparsely-Gated Mixture of Experts, for massively increasing the parameter count of language models

LLMDataHub. A quick guide (especially) for trending instruction finetuning datasets

CEPE. [ACL 2024] Long-Context Language Modeling with Parallel Encodings

flow_tutorials. Public tutorials of using Flow Forecast for forecasting and classifying time series data

flow-forecast. Deep learning PyTorch library for time series forecasting, classification, and anomaly detection (originally for flood forecasting).

alignment-handbook. Robust recipes to align language models with human and AI preferences

coral-pytorch. CORAL and CORN implementations for ordinal regression with deep neural networks.

UniRepLKNet. [CVPR 2024 & TPAMI 2025] UniRepLKNet

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

timeseries-clustering-vae. Variational Recurrent Autoencoder for timeseries clustering in pytorch

meta-prompting. Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffolding

minbpe. Minimal, clean code for the Byte Pair Encoding (BPE) algorithm commonly used in LLM tokenization.

tiktoken. tiktoken is a fast BPE tokeniser for use with OpenAI's models.

LongMamba. Some preliminary explorations of Mamba's context scaling.

axolotl. Go ahead and axolotl questions

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

contrastors. Train Models Contrastively in Pytorch

OLMo. Modeling, training, eval, and inference code for OLMo

MRL. Code repository for the paper - "Matryoshka Representation Learning"

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

RL-Adventure. Pytorch Implementation of DQN / DDQN / Prioritized replay/ noisy networks/ distributional values/ Rainbow/ hierarchical RL

Deep-reinforcement-learning-with-pytorch. PyTorch implementation of DQN, AC, ACER, A2C, A3C, PG, DDPG, TRPO, PPO, SAC, TD3 and ....

pytorch-transformer-ts. Repository of Transformer based PyTorch Time Series Models

s4. Structured state space sequence models

MASTER. This is the official code and supplementary materials for our AAAI-2024 paper: MASTER: Market-Guided Stock Transformer for Stock Price Forecasting. MASTER is a stock transformer for stock price forecasting, which models the momentary and cross-time stock correlation and guide feature selection with market information.

mamba-hf. Implementation of the Mamba SSM with hf_integration.

mamba. Mamba SSM architecture

Time-Series-Library. A Library for Advanced Deep Time Series Models for General Time Series Analysis.

TSMOM. Replication of Time Series Momentum strategy by Moskowtiz, Ooi, Pedersen, 2011.

alpaca_farm. A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

x-transformers. A concise but complete full-attention transformer with a set of promising experimental features from various papers

soft-moe-pytorch. Implementation of Soft MoE, proposed by Brain's Vision team, in Pytorch

TimeSeries. Implementation of deep learning models for time series in PyTorch.

DeepAR-pytorch. Python

llama-cookbook. Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started with Inference, Fine-Tuning, RAG. We also show you how to solve end to end problems using Llama model family and using them on various provider services

DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books. This jupyter notebook is used to demonstrate our recent work, "DeepLOB: Deep Convolutional Neural Networks for Limit Order Books", published in IEEE Transactions on Singal Processing. We use FI-2010 dataset and present how model architecture is constructed here. The FI-2010 is publicly avilable and interested readers can check out their paper.

Limit-Orderbook. Limit Order Book for high-frequency trading (HFT) strategies using data science approaches

PowerfulPromptFT. [NeurIPS 2023 Main Track] This is the repository for the paper titled "Don’t Stop Pretraining? Make Prompt-based Fine-tuning Powerful Learner"

xgen. Salesforce open-source LLMs with 8k sequence length.

neuralforecast. Scalable and user friendly neural :brain: forecasting algorithms.

vector-quantize-pytorch. Vector (and Scalar) Quantization, in Pytorch

PatchTST. An offical implementation of PatchTST: "A Time Series is Worth 64 Words: Long-term Forecasting with Transformers." (ICLR 2023) https://arxiv.org/abs/2211.14730

sd-scripts. Python

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

toolformer-pytorch. Implementation of Toolformer, Language Models That Can Use Tools, by MetaAI

textgen. Open-source desktop app for local LLMs. Text, vision, tool-calling, OpenAI/Anthropic-compatible API. 100% private.

peft. 🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

LLaMA-Adapter. [ICLR 2024] Fine-tuning LLaMA to follow Instructions within 1 Hour and 1.2M Parameters

GPT-4-LLM. Instruction Tuning with GPT-4

baize-chatbot. Let ChatGPT teach your own chatbot in hours with a single GPU!

FastChat. An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

chatgpt-retrieval-plugin. The ChatGPT Retrieval Plugin lets you easily find personal or work documents by asking questions in natural language.

dolly. Databricks’ Dolly, a large language model trained on the Databricks Machine Learning Platform

openai-cookbook. Examples and guides for using the OpenAI API

ai-pdf-chatbot-langchain. AI PDF chatbot agent built with LangChain & LangGraph

evals. Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.

pyllama. LLaMA: Open and Efficient Foundation Language Models

self-instruct. Aligning pretrained language models with instruction data generated by themselves.

stanford_alpaca. Code and documentation to train Stanford's Alpaca models, and generate the data.

lora. Using Low-rank adaptation to quickly fine-tune diffusion models.

llama. Inference code for Llama models

Prompt-Engineering-Guide. 🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.