将冰冷的离别化为温暖的 Skill，欢迎加入数字生命1.0！Transforming cold farewells into warm skills? It's giving rebirth era. Welcome to Digital Life 1.0. 🫶

The Feldera Incremental Computation Engine

Apache Paimon is a lake format that enables building a Realtime Lakehouse Architecture with Flink and Spark for both streaming and batch operations.

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.

"🐈 nanobot: The Ultra-Lightweight Personal AI Agent"

[VLDB'26] This repository provides a DuckDB implementation of RPT+, following a Yannakakis-style execution framework.

An extensible, state of the art columnar file format. Formerly at @spiraldb, now an Incubation Stage project at LFAI&Data, part of the Linux Foundation.

Algorithm powering the For You feed on X

High-performance data engine for AI and multimodal workloads. Process images, audio, video, and structured data at any scale

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

The Official Implementation of Ada-KV [NeurIPS 2025]

The evaluation framework for training-free sparse attention in LLMs

Turn GitHub Copilot into OpenAI/Anthropic API compatible server. Usable with Claude Code!

AIPex: AI browser automation assistant, no migration and privacy first. Alternative to Manus Browser Operator、 Claude Chrome and Agent Browser

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

LLM KV cache compression made easy

Efficient and easy multi-instance LLM serving

Cost-efficient and pluggable Infrastructure components for GenAI inference

KV cache store for distributed LLM inference

NVIDIA Inference Xfer Library (NIXL)

A Datacenter Scale Distributed Inference Serving Framework

FlashInfer: Kernel Library for LLM Serving

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

Large Language Model (LLM) Systems Paper List

A high-performance inference system for large language models, designed for production environments.

Hopsworks - Data-Intensive AI platform with a Feature Store

A low-latency & high-throughput serving engine for LLMs

This is RonDB, a distribution of NDB Cluster developed and used by Hopsworks AB. It also contains development branches of RonDB.