Python Repositories
[CVPR 2026 (Highlight)] Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction
WorldEngine: Towards the Era of Post-Training for Physical AI
🖥 Neural Computers' Data Engine
淘宝算法逆向，淘宝api，websockets自动运营，淘宝AI Agent基座
Production-ready toolkit for evaluating, monitoring, and ensuring safety of LLM deployments. Hallucination detection, bias evaluation, feedback loops, and production readiness assessment.
Code for the DISCO model: General Multimodal Protein Design Enables DNA-Encoding of Chemistry
A lightweight, prompt-driven MCP web research server for high-quality LLM powered information extraction.
No description available.
Multi-Agent DPO Data Synthesis Factory — 多智能体偏好训练数据自动合成框架 | 红队攻击 → 多persona审核 → 终审裁决 → DPO偏好对
PDF traditional-to-simplified Chinese conversion kit with layout-preserving scripts, Codex skill, and example source/output PDFs.
No description available.
MOSS-TTS-Nano is an open-source multilingual tiny speech generation model from MOSI.AI and the OpenMOSS team. With only 0.1B parameters, it is designed for realtime speech generation, can run directly on CPU without a GPU, and keeps the deployment stack simple enough for local demos, web serving, and lightweight product integration.
Exact speculative decoding on Apple Silicon, powered by MLX.
Karpathy's LLM-Wiki vision, fully realized — wiki-centric full-lifecycle AI research platform powered by Claude Code
No description available.
A from-scratch Prefill/Decode disaggregation inference engine for LLMs
Bypass DPI with IP/TCP-Header manipulation
Lossless DFlash speculative decoding for MLX on Apple Silicon
Yet another reusable writing skill for Chinese technical documentation and product copy.
The Ultimate ""Token Saver""
No description available.
A Cinematic audio dubbing, Cloning and voice generation studio
ParseBench - A Document Parsing Benchmark for AI Agents
No description available.
An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale
ERNIE-Image is an open text-to-image generation model developed by the ERNIE-Image team at Baidu. It is built on a single-stream Diffusion Transformer (DiT), with only 8B DiT parameters, it reaches state-of-the-art performance among open-weight text-to-image models.
AI-powered video editor that turns raw footage and a creative brief into a polished ad using an ensemble of AI agents (Google Gemini + FFmpeg)
Production-grade MCP server giving Claude 27 security intelligence tools across 21 APIs — CVE lookup, EPSS scoring, CISA KEV, MITRE ATT&CK, Shodan, VirusTotal, and more.
No description available.
面相A股市场的量化交易的脚手架工具（轻量级），涵盖历史/实时数据查询、策略信号、触达等功能，可以自由进行因子开发/交易模块的扩展，适合快速上手量化开发。
Personal bookkeeping app - a QuickBooks 2003 Pro replacement, decompiled from the ashes of QBW32.EXE. Free for personal and enterprise use.
Scaling Autonomous Research in Medical Image Segmentation
Reference code for the Meta-Harness paper.
No description available.
A feed-forward 3D foundation model for reconstructing scenes from streaming data
冰冷的钱就这样流进我的口袋-游资（UZI）Skills — 51位投资大佬帮你看盘 · 22维数据 × 180条量化规则 × 17种机构分析方法 · A股/港股/美股
No description available.
A minimal LLM-powered zero-day vulnerability scanner by AISLE.
No description available.
This is the official code repo for DiT4DiT, a Vision-Action-Model (VAM) framework that combines video generation model with flow-matching-based action prediction for generalizable robotic manipulation.
MOSS-Audio is an open-source foundation model for unified audio understanding, enabling speech, sound, music, captioning, QA, and reasoning in real-world scenarios.
No description available.
No description available.
One brain, many harnesses. Portable .agent/ folder (memory + skills + protocols) that plugs into Claude Code, Cursor, Windsurf, OpenCode, OpenClaw, Hermes, or DIY Python — and keeps its knowledge when you switch.
微信收藏可视化 Claude Code Skill — 从加密 DB 到交互式 HTML 报告的端到端管线
Hardware hacker’s flying probe automation stack for agent-driven target discovery, microscope mapping, safety-monitored CNC motion, probe review, and controlled pin probing.
Detect, count, and track truck traffic on any highway on Earth using nothing but free Sentinel-2 imagery and a browser.
No description available.
A world engine where AI agents live autonomously — physical rules, information asymmetry, any agent can plug in. Define scenarios in YAML, watch stories emerge.
Sync Apple iCloud Photos with Synology Diskstation
Self-healing browser harness that enables LLMs to complete any task.
A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.
LIDARLearn: A Unified Deep Learning Library for 3D Point Cloud Classification, Segmentation, and Self-Supervised Representation Learning
Production-Ready RAG with Structure-Aware Reasoning
Distributed ML training across Apple Silicon Macs
Tree-based speculative decoding for Apple Silicon (MLX). ~10-15% faster than DFlash on code, ~1.5x over autoregressive. First MLX port with custom Metal kernels for hybrid model support.
Local simulator & replay lab for XChat bots — debug webhooks, replay events, test E2EE flows without burning API credits
Causal Digital Twin for Marketing at Scale · Predict any marketing decision before you spend a dollar.
No description available.
An agent capable of self-evolving and dynamically hardening security