Tongyi Deep Research, the Leading Open-source Deep Research Agent

Convert Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF to clean Markdown. Built in Rust, with Node.js and Python bindings.

End-to-End Vision-Language Pretraining Without Negatives

The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/

一个基于多模态向量模型及视觉多模态模型构建的图片搜索引擎&管理系统，实现精准的以文搜文，文搜图、以图搜图多种智能检索方式。An image search engine management system built upon multimodal vector models and visual multimodal models, implementing multiple intelli…

Convert any file to LLM-ready Markdown - MCP Server + Claude Code Skill

World's first open-source, agentic video production system. 12 production pipelines, 100+ tools, 700+ agent skill and production-knowledge files. Turn your AI coding assistant into a full video pro…

Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.

📑 PageIndex: Document Index for Vectorless, Reasoning-based RAG

《深入理解 AI Agent：设计原理与工程实践》（李博杰 著）开源主仓库：全书正文、编译版 PDF 与按章配套代码

A new SOTA for RAG — an original retrieval architecture and an open-source knowledge base for humans and agents.

TreeRAG: Hierarchical RAG platform that indexes dense PDFs into logical trees for precise retrieval, 100% traceability, and LLM-guided deep traversal. Production-ready with enhanced security and st…

[ECCV 26] FSDC-DETR

Self-Evolving Vision-Language Models for Image Quality Assessment via Voting and Ranking (ICLR 2026)

Self-supervised learning for spatial perception

A blazing-fast, Rust-based code analysis CLI powered by tree-sitter. Analyze code structure, track dependencies, and explore call graphs across multiple languages. Optimized for AI coding agents

Sub2API 一站式开源中转服务，让 Claude、Openai 、Gemini、Grok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。

📚 《Deep Agents 实战》—— LangChain 官方大使出品，基于 LangChain / LangGraph 生态，从零构建生产级 AI Agent 的完整指南

YOLOE: Real-Time Seeing Anything [ICCV 2025]

High-Performance Face Recognition Framework (models trained on MS1MV2) | In PyTorch >> ONNX Runtime Inference

Pixel-in-Pixel Net: Towards Efficient Facial Landmark Detection in the Wild | PyTorch & ONNX Runtime Inference

Production-grade engineering skills for AI coding agents.

The official repository of Qwen-VLA

UniFace: A Unified Face Analysis Library for Python | Detection, alignment, landmarks, face-mesh, recognition, parsing, gaze, attributes and anti-spoofing under one API.

Efficient Universal Perception Encoder: a single on-device vision encoder with versatile representations that match or exceed specialized experts across multiple task domains.

[CVPR 2026] A Closer Look at Cross-Domain Few-Shot Object Detection: Fine-Tuning Matters and Parallel Decoder Helps