Stars
A Distributed, Fault-Tolerant Message Queue from Scratch. Inspired by Apache Kafka
1st Place Team Crane: @aswinkumar1999 @rathull @kyolebu
KernelBench: Can LLMs Write GPU Kernels? - Benchmark with Torch -> CUDA (+ more DSLs)
Intelligent automation and multi-agent orchestration for Claude Code
KAI Scheduler is an open source Kubernetes Native scheduler for AI workloads at large scale
CUDA Python: Performance meets Productivity
SGLang is a fast serving framework for large language models and vision language models.
Distributed Compiler based on Triton for Parallel Systems
Simple, complete, correct, optimal and industrial quality solutions for MIT 6.824 distributed systems course
A uniform interface to run deep learning models from multiple frameworks
LevelCache is an ephemeral embedded cache with TTL support built on top of LevelDB.
Python tool for converting files and office documents to Markdown.
A lightweight, powerful framework for multi-agent workflows
ACI.dev is the open source tool-calling platform that hooks up 600+ tools into any agentic IDE or custom AI agent through direct function calling or a unified MCP server. The birthplace of VibeOps.
A Go implementation of the Model Context Protocol (MCP), enabling seamless integration between LLM applications and external data sources and tools.
An open protocol enabling communication and interoperability between opaque agentic applications.
The book "Performance Analysis and Tuning on Modern CPU"
Implementing the 4 agentic patterns from scratch
Deploy your agentic worfklows to production
Flexible and powerful framework for managing multiple AI agents and handling complex conversations
Class materials for a distributed systems lecture series
A toolkit to run Ray applications on Kubernetes
vLLM’s reference system for K8S-native cluster-wide deployment with community-driven performance optimization
A lightweight data processing framework built on DuckDB and 3FS.