Maru
/

Getting Started

  • System Setup
  • Installation
  • Quick Start
  • Examples
    • vLLM Examples
    • LMCache
      • Disaggregated Prefill
      • P2P KV Cache Sharing
    • Dynamo Examples

Design Docs

  • Architecture Overview
    • Maru Handler
    • Maru Server
    • Maru Resource Manager
    • Maru FS
  • Memory Model
  • KV Cache Management
  • Consistency and Safety
  • CPU/CXL L1 and SSD L2 (Proposal)

Integration

  • LMCache
  • vLLM
  • NVIDIA Dynamo
  • SGLang HiCache

API Reference

  • Python API
  • Configuration
  • Handler Plugins
  • Resource Manager Admin (maru-rm-ctl)

Performance

  • LMCache Benchmark

Contact

  • Contact Us
  1. Maru /
  2. Examples

Examples#

Start with Quick Start for single-host or multi-host KV sharing. The examples below cover vLLM, LMCache, and Dynamo serving configurations.

  • vLLM Examples
  • LMCache
    • Disaggregated Prefill
    • P2P KV Cache Sharing
  • Dynamo Examples
Previous
Quick Start
Next
vLLM Examples

© 2026 XCENA Inc. | All Rights Reserved

Made with Sphinx and Shibuya theme.