BIRD-CRITIC 1.0: Can Large Language Models Solve USER SQL Issues in Real-World Database Applications?

A curated list of resources on graph-based retrieval-augmented generation (GraphRAG) for customized large language models.

Open-Tax is an AI-powered cloud platform transforming tax compliance through automated data integration, real-time anomaly detection, and multi-jurisdictional reconciliation.

Uncommon Objects in 3D dataset

PVPAI LLM 🔥The First Open-Source DeFAI Large Language Model Powered by DeepSeek.

A curated list of papers, code and resources pertaining to image composition/compositing or object insertion/addition/compositing, which aims to generate realistic composite image.

Manuscript is a revolutionary blockchain data streaming framework. With Manuscript, you can seamlessly integrate on-chain and off-chain data into target data storage for unrestricted querying and a…

Hallo3: Highly Dynamic and Realistic Portrait Image Animation with Diffusion Transformer Networks

Align Anything: Training All-modality Model with Feedback

Some tools for cloud developers

A Speech-to-Text Input Method For Windows

Eko (Eko Keeps Operating) - Build Production-ready Agentic Workflow with Natural Language - eko.fellou.ai

A highly optimized LLM inference acceleration engine for Llama and its variants.

Curated selection of technical publications on the solana blockchain, with a focus on DeFi, infrastructure, MEV, and fee (free) markets.

Urani's first MEV agent template. It listens for batches, parses them, checks for P2P matches, and spins threads for best quotes through arbitrage in different AMMs.

The official implementation of MARS: Unleashing the Power of Variance Reduction for Training Large Models

以开源为核心的IDaas/IAM平台，用于管理企业内员工账号、权限、身份认证、应用访问，帮助整合部署在本地或云端的内部办公系统、业务系统及三方 SaaS 系统的所有身份，实现一个账号打通所有应用的服务。

Run AI models end-to-end encrypted.

[LCLR 2025 Oral] TANGO: Co-Speech Gesture Video Reenactment with Hierarchical Audio-Motion Embedding and Diffusion Interpolation

数据流引擎是一款面向数据集成、数据同步、数据交换、数据共享、任务配置、任务调度的底层数据驱动引擎。数据流引擎采用管执分离、多流层、插件库等体系应对大规模数据任务、数据高频上报、数据高频采集、异构数据兼容的实际数据问题。

数据标注是一款专门对文本数据进行处理和标注的工具，通过简化快捷的文本标注流程和动态的算法反馈，支持用户快速标注关键词并能通过算法持续减少人工标注的成本和时间。数据标注的过程先由人工标注构建基础，再由自动标注反哺人工标注，最后由人工标注进行纠偏，从而大幅度提高标注的精准度和高效性。数据标注需要依赖开源的数字底座进行人员岗位管控。

AML end to end system

Unified KV Cache Compression Methods for Auto-Regressive Models

One-stop data intelligence agent, providing insights from all mainstream data formats in a single dialogue box, including documents, databases, business systems, and images.一站式数据智能体，一个对话框提供所有主流格式数据…

"LightRAG: Simple and Fast Retrieval-Augmented Generation"

The official repository for paper "Tora: Trajectory-oriented Diffusion Transformer for Video Generation"

The first open autoregressive foundational video AI model.