基于 Claude Code 的长篇网文辅助创作系统，解决 AI 写作中的「遗忘」和「幻觉」问题，支持 200 万字量级 连载创作。

A Blender extension for automated, editable, and inter-shot consistent 3D storyboard production.

Implementation of the AAAI 2026 paper "StoryBox: Collaborative Multi-Agent Simulation for Hybrid Bottom-Up Long-Form Story Generation Using Large Language Models".

ConStory-Bench (ACL 26')

collection of diffusion model papers categorized by their subareas

Official Implementation for *PaCo-RL: Advancing Reinforcement Learning for Consistent Image Generation with Pairwise Reward Modeling*

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

原汁原昧 Claude Code 可运行,可构建, 可调试版; 生产级工程化, 企业级可靠性; 安全无毒, 内存泄露修复

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Harness Engineering 学习指南 — 从概念理解到独立实践的深度学习档案

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

The Cradle framework is a first attempt at General Computer Control (GCC). Cradle supports agents to ace any computer task by enabling strong reasoning abilities, self-improvment, and skill curatio…

OpenClaw-RL: Train any agent simply by talking

Implement a reasoning LLM in PyTorch from scratch, step by step

Code for "Adversarial Illusions in Multi-Modal Embeddings"

Code for "Defending LVLMs Against Vision Attacks through Partial-Perception Supervision"

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

Build AI agents from first principles using a local LLM - no frameworks, no cloud APIs, no hidden reasoning.

A scalable, end-to-end training pipeline for general-purpose agents

The Open Cookbook for Top-Tier Code Large Language Model

OpenSource Claude Cowork. A desktop AI assistant that helps you with programming, file management, and any task you can describe.

"RAG-Anything: All-in-One RAG Framework"

Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"

Witness the aha moment of VLM with less than $3.

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

pytorch implementation of Manifold-constrained Hyper Connection (mHC)

DeepSeek Native Sparse Attention pytorch implementation

MAG-SQL: Multi-Agent Generative Approach with Soft Schema Linking and Iterative Sub-SQL Refinement for Text-to-SQL