Haiyang Wang

tmax. Training terminal-agents

agents-last-exam. Agents' Last Exam

SWE-CI. SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration

DeepSpec. DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

cc-download. claude code install and update on windows OS

Qwen-AgentWorld. Qwen-AgentWorld: Language World Models for General Agents

mcore-bridge. MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transformers — with support for 300+ large language models (Qwen3-Next, GLM-5.2, Deepseek-V4, MiniMax-2.7, ...) and 200+ multimodal large models (Qwen3.5, Qwen3-Omni, Gemma4, ...).

frontier-swe. FrontierSWE is an ultra long-horizon coding agent benchmark that tests implementation, performance eng and ML research

dgm. Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents

SEAL. Self-Adapting Language Models

skills. Skills for Real Engineers. Straight from my .agents directory.

mobilegym. MobileGym: A Verifiable and Highly Parallel Simulation Platform for Mobile GUI Agent Research · 浏览器里运行的安卓模拟器 · Browser-hosted Android Simulator · Verifiable Evaluation · Scalable Online RL Training

OpenHarnessExtended. Python

Claw-Anything. official implementation of "Claw-Anything: Benchmarking Always-On Personal Assistants with Broader Access to the User's Digital World"

Awesome-Issue-Resolution. Advances and Frontiers of LLM-based Issue Resolution in Software Engineering A Comprehensive Survey

dexjoco. Python

OpenHarness. "OpenHarness: Open Agent Harness with a Built-in Personal Agent--Ohmo!"

ClawEnvKit. Open-source Environment toolkit of claw-like agents, support task/harness generation and evaluation

ClawGUI. Build, Evaluate, and Deploy GUI Agents — online RL training, standardized benchmarks, and real-device deployment in one framework.

cli. Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills.

NemoClaw. Run agents like Hermes, LangChain Deep Agents, and OpenClaw more securely inside NVIDIA OpenShell with managed inference

hermes-agent. The agent that grows with you

claw-eval. Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.

WildClawBench. An in-the-wild benchmark for AI agents in the OpenClaw Environment.

public-image-mirror. 很多镜像都在国外。比如 gcr 。国内下载很慢，需要加速。致力于提供连接全世界的稳定可靠安全的容器镜像服务。

awesome-autoresearch. Curated list of AutoResearch use cases with optimization traces and open source implementations

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

claude-relay-service. CRS-自建Claude Code镜像，一站式开源中转服务，让 Claude、OpenAI、Gemini、Droid 订阅统一接入，支持拼车共享，更高效分摊成本，原生工具无缝使用。

ROLL. An Efficient and User-Friendly Scaling Library for Reinforcement Learning with Large Language Models

claude-code-design-guide. From Early Internet Design Patterns to AI Agent Implementation — A Deep Dive into Claude Code for Developers

CatchMe. "CatchMe: Make Your AI Agents Truly Personal"

claw-bench. The Definitive AI Agent Benchmark

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

ClawTeam. "ClawTeam: Agent Swarm Intelligence" (One Command → Full Automation)

mercury. 🪽 Mercury — There are many claws, but this one is mine.

agency-agents. A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.

skill. PinchBench is a benchmarking system for evaluating LLM models as OpenClaw coding agents. Made with 🦀 by the humans at https://kilo.ai

openfang. Open-source Agent Operating System

eigent. Eigent: The Open Source Cowork Desktop - Local and Free Alternative to Claude Cowork and Codex

symphony. Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents.

nanobot. Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps

awesome-openclaw. Python

awesome-openclaw-usecases. A community collection of OpenClaw use cases for making life easier.

awesome-openclaw-skills. The awesome collection of OpenClaw skills. 5,400+ skills filtered and categorized from the official OpenClaw Skills Registry.🦞

v1. Open-source multi-agent platform that structures software engineering as a coordinated team of specialized AI agents. It enforces strict role separation, isolated execution sandboxes, structured communication, and secrets management.

AgentTeams. An open-source Collaborative Multi-Agent OS for transparent, human-in-the-loop task coordination via Matrix rooms.

agents. Multi-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLI

ScaleSWE.

slime. slime is an LLM post-training framework for RL Scaling.

nanoclaw. A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK

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

CLI-Gym. Official Implementation of "CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion"

FeatureBench. [ICLR 2026] Official Implementation of "FeatureBench: Benchmarking Agentic Coding for Complex Feature Development"

Open-AgentRL. RLAnything (ICML 2026) & AutoTool (ICML 2026), DemyAgent: Open-Source RL for LLMs and Agentic Scenarios

voltagent. AI Agent Engineering Platform built on an Open Source TypeScript AI Agent Framework

swe-factory. [FSE'2026] SWE-Factory: Your Automated Factory for Issue Resolution Training Data and Evaluation Benchmarks

opencode. The open source coding agent.

Engram. Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models

IQuest-Coder-V1. Python

benchmarks. Evaluation harness for OpenHands V1.

UniSplat. [ICLR 2026] Official Implementation of "UniSplat: Unified Spatio-Temporal Fusion via 3D Latent Scaffolds for Dynamic Driving Scene Reconstruction""

clash-for-linux-install. 😼 优雅地使用基于 clash/mihomo 的代理环境

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

static-analysis. ⚙️ A curated list of static analysis (SAST) tools and linters for all programming languages, config files, build tools, and more. The focus is on tools which improve code quality.

Awesome-Self-Evolving-Agents. [Survey] A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems

DeepCode. "DeepCode: Open Agentic Coding (Paper2Code & Text2Web & Text2Backend)"

context7. Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors

mini-swe-agent. The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

polars. Extremely fast Query Engine for DataFrames, written in Rust

SWE-Swiss. SWE-Swiss: A Multi-Task Fine-Tuning and RL Recipe for High-Performance Issue Resolution

CURE. [NeurIPS 2025 Spotlight] Co-Evolving LLM Coder and Unit Tester via Reinforcement Learning

R-Zero. [ICLR2026] codes for R-Zero: Self-Evolving Reasoning LLM from Zero Data (https://www.arxiv.org/pdf/2508.05004)

deepwiki-open. Open Source DeepWiki: AI-Powered Wiki Generator for GitHub/Gitlab/Bitbucket Repositories. Join the discord: https://discord.gg/gMwThUMeme

qwen-code. An open-source AI coding agent that lives in your terminal.

gemini-cli. An open-source AI agent that brings the power of Gemini directly into your terminal.

learn-claude-code. Bash is all you need - A nano claude code–like 「agent harness」, built from 0 to 1

Kode-CLI. Kode CLI — Design for post-human workflows. One unit agent for every human & computer task.

ml-diffucoder. DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation

AI-Coding-Style-Guides. A set of coding style guidelines for Vibe Coding or SWE-Agents that maximize efficiency and improve human readability.

trae-agent. Trae Agent is an LLM-based agent for general purpose software engineering tasks.

KernelBench. KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

servers. Model Context Protocol Servers

swe-rl. [NeurIPS'25] Official codebase for "SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution"

Awesome-Inference-Time-Scaling. Paper List of Inference/Test Time Scaling/Computing

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

ms-swift. Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).

discrete-diffusion-papers. A collection of papers on discrete diffusion models

openevolve. Open-source implementation of AlphaEvolve

codex. Lightweight coding agent that runs in your terminal

awesome-hallucination-detection. List of papers on hallucination detection in LLMs.

hallucination-leaderboard. Leaderboard Comparing LLM Performance at Producing Hallucinations when Summarizing Short Documents

Code-Agent-Survey. A survey of Code Agents / Foundation Models for improving development productivity. Become 10x SWE, MLE, etc.

code-act. Official Repo for ICML 2024 paper "Executable Code Actions Elicit Better LLM Agents" by Xingyao Wang, Yangyi Chen, Lifan Yuan, Yizhe Zhang, Yunzhu Li, Hao Peng, Heng Ji.

rllm. Democratizing Reinforcement Learning for LLMs

Search-R1. Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL

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

Search-o1. 🔍 Search-o1: Agentic Search-Enhanced Large Reasoning Models [EMNLP 2025]

open-deep-research. An open source deep research clone. AI Agent that reasons large amounts of web data extracted with Firecrawl

augment-swebench-agent. The #1 open-source SWE-bench Verified implementation

OpenHands. 🙌 OpenHands: AI-Driven Development

awesome-language-agents. List of language agents based on paper "Cognitive Architectures for Language Agents"

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

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]

Agent4SE-Paper-List. Repository for the paper "Large Language Model-Based Agents for Software Engineering: A Survey". Keep updating.

Qwen2.5-Omni. Qwen2.5-Omni is an end-to-end multimodal model by Qwen team at Alibaba Cloud, capable of understanding text, audio, vision, video, and performing real-time speech generation.

DeepSeek-Coder-V2. DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

awesome-deepseek-coder. A curated list of open-source projects related to DeepSeek Coder

bigcode-dataset. Jupyter Notebook

SWE-bench. SWE-bench: Can Language Models Resolve Real-world Github Issues?

Awesome-Agent-Papers. [Up-to-date] Large Language Model Agent: A Survey on Methodology, Applications and Challenges

Awesome-LLM. Awesome-LLM: a curated list of Large Language Model

Awesome-Knowledge-Distillation-of-LLMs. This repository collects papers for "A Survey on Knowledge Distillation of Large Language Models". We break down KD into Knowledge Elicitation and Distillation Algorithms, and explore the Skill & Vertical Distillation of LLMs.

self-correction-llm-papers. This is a collection of research papers for Self-Correcting Large Language Models with Automated Feedback.

fastText. Library for fast text representation and classification.

cc_net. Tools to download and cleanup Common Crawl data

frontier-evals. OpenAI Frontier Evals

openinterpreter. A coding agent for open models like Kimi K3

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

owl. 🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation

VICT. [CVPR 2025] Test-Time Visual In-Context Tuning

repo-level-codegen-papers. Repo-Level Code generation papers

Awesome-Code-LLM. [TMLR] A curated list of language modeling researches for code (and other software engineering activities), plus related datasets.

Awesome-LLM-Reasoning. From Chain-of-Thought prompting to OpenAI o1 and DeepSeek-R1 🍓

TinyZero. Minimal reproduction of DeepSeek R1-Zero

SPIN. The official implementation of Self-Play Fine-Tuning (SPIN)

Awesome-LLM-Post-training. Awesome Reasoning LLM Tutorial/Survey/Guide

GoT. Official repository of "GoT: Unleashing Reasoning Capability of Multimodal Large Language Model for Visual Generation and Editing"

Mani-GS. [CVPR' 2025'] Mani-GS: Gaussian Splatting Manipulation with Triangular Mesh

camel. 🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

UFO. [NeurIPS2025 Spotlight 🔥 ] Official implementation of 🛸 "UFO: A Unified Approach to Fine-grained Visual Perception via Open-ended Language Interface"

Qwen-Agent. Agent framework and applications built upon Qwen>=3.0, featuring Function Calling, MCP, Code Interpreter, RAG, Chrome extension, etc.

native-sparse-attention. 🐳 Efficient Triton implementations for "Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention"

DeepEP. DeepEP: an efficient expert-parallel communication library

open-infra-index. Production-tested AI infrastructure tools for efficient AGI development and community-driven innovation

FlashMLA. FlashMLA: Efficient Multi-head Latent Attention Kernels

Open-Reasoner-Zero. Official Repo for Open-Reasoner-Zero

MoBA. MoBA: Mixture of Block Attention for Long-Context LLMs

thinking-in-space. Official repo and evaluation implementation of VSI-Bench

leetcode-master. 《代码随想录》LeetCode 刷题攻略：200道经典题目刷题顺序，共60w字的详细图解，视频难点剖析，50余张思维导图，支持C++，Java，Python，Go，JavaScript等多语言版本，从此算法学习不再迷茫！🔥🔥 来看看，你会发现相见恨晚！🚀

DeepSeek-Math. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Qwen. The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

open-r1. Fully open reproduction of DeepSeek-R1

DeepSeek-R1.

OpenAgents. [COLM 2024] OpenAgents: An Open Platform for Language Agents in the Wild

awesome-llm-powered-agent. Awesome things about LLM-powered agents. Papers / Repos / Blogs / ...

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

MoE-Jetpack. [NeurIPS 24] MoE Jetpack: From Dense Checkpoints to Adaptive Mixture of Experts for Vision Tasks

Pyramid-Flow. [ICLR 2025] Pyramidal Flow Matching for Efficient Video Generative Modeling

TokenFormer. [ICLR2025 Spotlight🔥] Official Implementation of TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters

Book-Mathematical-Foundation-of-Reinforcement-Learning. This is the homepage of a new book entitled "Mathematical Foundations of Reinforcement Learning."

ToolBench. [ICLR'24 spotlight] An open platform for training, serving, and evaluating large language model for tool learning.

ToolLearningPapers.

quiet-star. Code for Quiet-STaR

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

trl. Train transformer language models with reinforcement learning.

direct-preference-optimization. Reference implementation for DPO (Direct Preference Optimization)

PaLM-rlhf-pytorch. Implementation of RLHF (Reinforcement Learning with Human Feedback) on top of the PaLM architecture. Basically ChatGPT but with PaLM

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

llama-moe. ⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)

RedPajama-Data. The RedPajama-Data repository contains code for preparing large datasets for training large language models.

gemma. Gemma open-weight LLM library, from Google DeepMind

LL3M. LL3M: Large Language and Multi-Modal Model in Jax

EasyLM. Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Flax.

maxtext. A simple, performant, and scalable Jax LLM!

gpt-jax. Jax/Flax rewrite of Karpathy's nanoGPT

MambaOut. MambaOut: Do We Really Need Mamba for Vision? (CVPR 2025)

pykan. Kolmogorov Arnold Networks

ConTex-Human. [CVPR' 2024'] ConTex-Human: Free-View Rendering of Human from a Single Image with Texture-Consistent Synthesis

CoSSL. Official PyTorch Implementation of "CoSSL: Co-Learning of Representation and Classifier for Imbalanced Semi-Supervised Learning" (CVPR 2022)

latentsplat. [ECCV 2024] Implementation of latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction

BLIP. PyTorch code for BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

datacomp. DataComp: In search of the next generation of multimodal datasets

pythia. The hub for EleutherAI's work on interpretability and learning dynamics

open_clip. An open source implementation of CLIP.

FouriScale. Official implementation of FouriScale (ECCV2024)

Graphormer-GD. [ICLR 2023 notable top-5%] Rethinking the Expressive Power of GNNs via Graph Biconnectivity (official implementation)

VAR. [NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

gpt-neox. An implementation of model parallel autoregressive transformers on GPUs, based on the Megatron and DeepSpeed libraries

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

VIRL. (ECCV 2024) Code for V-IRL: Grounding Virtual Intelligence in Real Life

grok-1. Grok open release

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

LLoVi. Official implementation for "A Simple LLM Framework for Long-Range Video Question-Answering"

MoE-LLaVA. 【TMM 2025🔥】 Mixture-of-Experts for Large Vision-Language Models

SparK. [ICLR'23 Spotlight🔥] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"

GiT. [ECCV2024 Oral🔥] Official Implementation of "GiT: Towards Generalist Vision Transformer through Universal Language Interface"

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

gemma_pytorch. The official PyTorch implementation of Google's Gemma models

lm-evaluation-harness. A framework for few-shot evaluation of language models.

s4. Structured state space sequence models

mamba. Mamba SSM architecture

LoRA. Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

awesome-ChatGPT-repositories. A curated list of open source GitHub repositories related to ChatGPT, the OpenAI API, and Codex. Searchable via Claude Code skills.

lit-llama. Implementation of the LLaMA language model based on nanoGPT. Supports flash attention, Int8 and GPTQ 4bit quantization, LoRA and LLaMA-Adapter fine-tuning, pre-training. Apache 2.0-licensed.

picoGPT. An unnecessarily tiny implementation of GPT-2 in NumPy.

torchscale. Foundation Architecture for (M)LLMs

ScatterFormer. ScatterFormer: Efficient Voxel Transformer with Scattered Linear Attention (ECCV 2024)

bidiff. [CVPR'24] Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors

nanoGPT. The simplest, fastest repository for training/finetuning medium-sized GPTs.

HEDNet. HEDNet (NeurIPS 2023) & SAFDNet (CVPR 2024 Oral)