程序员延寿指南 | A programmer's guide to live longer

S3 vector database for LLM Agents and RAG.

Accelerating MoE with IO and Tile-aware Optimizations

[SIGMOD2026] Reveal Hidden Pitfalls and Navigate Next Generation of Vector Similarity Search with Task-Centric Benchmarks

ZeroSearch: Incentivize the Search Capability of LLMs without Searching

SigNoz is an open-source observability platform native to OpenTelemetry with logs, traces and metrics in a single application. An open-source alternative to DataDog, NewRelic, etc. 🔥 🖥. 👉 Open sour…

PDF craft can convert PDF files into various other formats. This project will focus on processing PDF files of scanned books.

TurboDiffusion: 100–200× Acceleration for Video Diffusion Models

Light Video Generation Inference Framework

Adamas: Hadamard Sparse Attention for Efficient Long-context Inference

Cookbook of SGLang - Recipe

DS SERVE: The Largest Open Vector Store over Pretain Data; A Framework for Efficient and Scalable Neural Retrieval

Tile-Based Runtime for Ultra-Low-Latency LLM Inference

Ada-ef — Adaptive efSearch for HNSW-based vector search

train a model on huchenfeng dataset

The ultimate training toolkit for finetuning diffusion models

A unified library for building, evaluating, and storing speculative decoding algorithms for LLM inference in vLLM

A framework for efficient model inference with omni-modality models

NVIDIA cuTile learn

Helpful kernel tutorials and examples for tile-based GPU programming

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

A comprehensive guide for beginners in the field of data management and artificial intelligence.

Code for the paper “Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling”

Milvus web documents and contents

Official inference repo for FLUX.2 models

Classic papers and resources on recommendation