Mooncake

About this project

A KVCache-centric Disaggregated Architecture for LLM Serving Paper Slides Traces Documentation Blog Slack as a rollout data-transfer backend for the fragmented, heterogeneous data moving between rollout and training in disaggregated RL. Blogs: [KVCache.AI, lmsys. — Aug 17, 2026: Mooncake is integrated into Speculators as a distributed backend for multi-node online training, efficiently moves hidden-state between vLLM inference workers and Speculators trainers through RDMA, eliminating the need for massive hidden-state storage in offline training. Benchmark on GB300 NVL72. — May 7, 2026: 🚀 vLLM officially features Mooncake Store — a deep dive into how…

GitHub

Reviews

Built by

kvcache-ai

Imported from GitHub · not yet claimed on GitPalace

View developer pageSign in with GitHub to claim

Maintain kvcache-ai/Mooncake? Claiming verifies admin access through your GitHub account and gives you control of this listing.

You might also like

openclaw

AI

The AI that really does things. Any OS. Any Platform. The lobster way. 🦞

★ New389K stars

by openclaw

superpowers

AI

An agentic skills framework & software development methodology that works.

★ New282.4K stars

by obra

hermes-agent

AI

The agent that grows with you

★ New242.5K stars

by NousResearch

n8n

AI

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host

★ New203.6K stars

by n8n-io