# DroidSpeak: KV Cache Sharing for Cross-LLM Communication and Multi-LLM Serving

arXiv: https://arxiv.org/abs/2411.02820 · HTML: https://mlsanity.com/paper/2411.02820/

## Key facts

- Cited by 35 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- First posted on arXiv 5 Nov 2024.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)
- Efficient LLM inference and serving / [Efficient reasoning and agent inference](https://mlsanity.com/inference/efficient-reasoning-and-agent-inference/)

## Built on by (most cited first, 30 of 35)

- [Cache-to-Cache: Direct Semantic Communication Between Large Language Models](https://mlsanity.com/paper/2510.03215/) (2025)
- [A Survey on LLM-based Multi-Agent System: Recent Advances and New Frontiers in Application](https://mlsanity.com/paper/2412.17481/) (2024)
- [KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems](https://mlsanity.com/paper/2510.12872/) (2025)
- [Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents](https://mlsanity.com/paper/2506.14852/) (2025)
- [KVComm: Enabling Efficient LLM Communication through Selective KV Sharing](https://mlsanity.com/paper/2510.03346/) (2025)
- [LRAgent: Efficient KV Cache Sharing for Multi-LoRA LLM Agents](https://mlsanity.com/paper/2602.01053/) (2026)
- [Multi-Agent Memory from a Computer Architecture Perspective: Visions and Challenges Ahead](https://mlsanity.com/paper/2603.10062/) (2026)
- [Optimizing Diversity and Quality through Base-Aligned Model Collaboration](https://mlsanity.com/paper/2511.05650/) (2025)
- [RelayCaching: Accelerating LLM Collaboration via Decoding KV Cache Reuse](https://mlsanity.com/paper/2603.13289/) (2026)
- [TokenDance: Scaling Multi-Agent LLM Serving via Collective KV Cache Sharing](https://mlsanity.com/paper/2604.03143/) (2026)
- [Batch Query Processing and Optimization for Agentic Workflows](https://mlsanity.com/paper/2509.02121/) (2025)
- [Sutradhara: An Intelligent Orchestrator-Engine Co-design for Tool-based Agentic Inference](https://mlsanity.com/paper/2601.12967/) (2026)
- [An experimental study of KV cache reuse strategies in chunk-level caching systems](https://mlsanity.com/paper/2603.20218/) (2026)
- [Semantic Cache Distillation: Efficient State Transfer via Reuse and Selective Patching](https://mlsanity.com/paper/2606.07684/) (2026)
- [PrefillShare: A Shared Prefill Module for KV Reuse in Multi-LLM Disaggregated Serving](https://mlsanity.com/paper/2602.12029/) (2026)
- [OxyGen: Unified KV Cache Management for VLA Inference under Multi-Task Parallelism](https://mlsanity.com/paper/2603.14371/) (2026)
- [PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents](https://mlsanity.com/paper/2605.19932/) (2026)
- [KV Cache Translation across Heterogeneous Large Language Models](https://mlsanity.com/paper/2607.28979/) (2026)
- [Generative Caching for Structurally Similar Prompts and Responses](https://mlsanity.com/paper/2511.17565/) (2025)
- [ForkKV: Scaling Multi-LoRA Agent Serving via Copy-on-Write Disaggregated KV Cache](https://mlsanity.com/paper/2604.06370/) (2026)
- [A Policy-Driven Runtime Layer for Agentic LLM Serving](https://mlsanity.com/paper/2605.27744/) (2026)
- [SwiftCache: Efficient LLM Serving for Multi-turn Conversations with Heterogeneous KV Cache Sharing](https://mlsanity.com/paper/2606.16135/) (2026)
- [Learning Agent Execution for KV-Cache Management in Agentic Serving](https://mlsanity.com/paper/2608.14624/) (2026)
- [PlanetServe: A Decentralized, Scalable, and Privacy-Preserving Overlay for Democratizing Large Language Model Serving](https://mlsanity.com/paper/2504.20101/) (2025)
- [QKVShare: Quantized KV-Cache Handoff for Multi-Agent On-Device LLMs](https://mlsanity.com/paper/2605.03884/) (2026)
- [TOPAS: Workflow-Aware Prefix-State Scheduling for Multi-Agent LLM Serving](https://mlsanity.com/paper/2608.25523/) (2026)
- [KVShareArena: KV-Cache Reuse Across Contexts and Model Checkpoints](https://mlsanity.com/paper/2609.10266/) (2026)
- [KV-Lingo: Learning KV-Cache Translators with Distillation](https://mlsanity.com/paper/2609.32610/) (2026)
- [CIFLEX: Contextual Instruction Flow for Sub-task Execution in Multi-Turn Interactions with a Single On-Device LLM](https://mlsanity.com/paper/2510.01239/) (2025)
- [Accelerating Language Model Workflows with Prompt Choreography](https://mlsanity.com/paper/2512.23049/) (2025)

## Abstract (arXiv)

Compound AI systems, such as agentic systems, are an emerging trend in large-scale enterprise settings, with multiple LLMs specialized for different users, tasks, and/or roles working together. In these scenarios, different models often process inputs that share the same context prefix. Although much work was done in the past to enable the reuse of prefix KV caches across inputs for a single model, how to enable one model to reuse the prefix KV caches of a different model remains an open question. We introduce DroidSpeak, the first distributed LLM inference system that enables KV cache reuse across distributed nodes running inference of different LLMs, so long as the LLMs have the same architecture. We present the first study that aims at understanding the impact of sharing KV caches across different LLMs, and if/when such sharing affects quality. Inspired by the findings, we present DroidSpeak, which selectively recomputes a few layers of the KV cache produced by another LLM and reuses the remaining layers, with negligible quality loss. Moreover, carefully pipelining the layer-wise re-computation and the loading of reused KV cache further improves the inference performance. Experiments on diverse datasets and model pairs demonstrate that DroidSpeak achieves up to 4x throughput improvement and about 3.1x faster prefill (time to first token), with negligible loss of quality in F1 scores, Rouge-L or code similarity score, compared to the baseline which does not allow any sharing across models.