# Marconi: Prefix Caching for the Era of Hybrid LLMs

arXiv: https://arxiv.org/abs/2411.19379 · HTML: https://mlsanity.com/paper/2411.19379/

## Key facts

- Cited by 31 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 58 references that are in the map.
- First posted on arXiv 28 Nov 2024.

## Builds on (most cited first, 30 of 39)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [RWKV: Reinventing RNNs for the Transformer Era](https://mlsanity.com/paper/2305.13048/) (2023)
- [Retentive Network: A Successor to Transformer for Large Language Models](https://mlsanity.com/paper/2307.08621/) (2023)
- [Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve](https://mlsanity.com/paper/2403.02310/) (2024)
- [FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision](https://mlsanity.com/paper/2407.08608/) (2024)
- [Gated Linear Attention Transformers with Hardware-Efficient Training](https://mlsanity.com/paper/2312.06635/) (2023)
- [LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset](https://mlsanity.com/paper/2309.11998/) (2023)
- [Jamba: A Hybrid Transformer-Mamba Language Model](https://mlsanity.com/paper/2403.19887/) (2024)
- [Parallelizing Linear Transformers with the Delta Rule over Sequence Length](https://mlsanity.com/paper/2406.06484/) (2024)
- [Learning to (Learn at Test Time): RNNs with Expressive Hidden States](https://mlsanity.com/paper/2407.04620/) (2024)
- [xLSTM: Extended Long Short-Term Memory](https://mlsanity.com/paper/2405.04517/) (2024)
- [Prompt Cache: Modular Attention Reuse for Low-Latency Inference](https://mlsanity.com/paper/2311.04934/) (2023)
- [InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management](https://mlsanity.com/paper/2406.19707/) (2024)
- [Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models](https://mlsanity.com/paper/2402.19427/) (2024)
- [LooGLE: Can Long-Context Language Models Understand Long Contexts?](https://mlsanity.com/paper/2311.04939/) (2023)
- [An Empirical Study of Mamba-based Language Models](https://mlsanity.com/paper/2406.07887/) (2024)
- [Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling](https://mlsanity.com/paper/2406.07522/) (2024)

## Built on by (most cited first, 30 of 31)

- [SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning](https://mlsanity.com/paper/2504.07891/) (2025)
- [Jenga: Effective Memory Management for Serving LLM with Heterogeneity](https://mlsanity.com/paper/2503.18292/) (2025)
- [METIS: Fast Quality-Aware RAG Systems with Configuration Adaptation](https://mlsanity.com/paper/2412.10543/) (2024)
- [Taming the Titans: A Survey of Efficient LLM Inference Serving](https://mlsanity.com/paper/2504.19720/) (2025)
- [Seesaw: High-throughput LLM Inference via Model Re-sharding](https://mlsanity.com/paper/2503.06433/) (2025)
- [Gated KalmaNet: A Fading Memory Layer Through Test-Time Ridge Regression](https://mlsanity.com/paper/2511.21016/) (2025)
- [ContextPilot: Fast Long-Context Inference via Context Reuse](https://mlsanity.com/paper/2511.03475/) (2025)
- [Sparse Prefix Caching for Hybrid and Recurrent LLM Serving](https://mlsanity.com/paper/2605.05219/) (2026)
- [VibeServe: Can AI Agents Build Bespoke LLM Serving Systems?](https://mlsanity.com/paper/2605.06068/) (2026)
- [Irminsul: MLA-Native Position-Independent Caching for Agentic LLM Serving](https://mlsanity.com/paper/2605.05696/) (2026)
- [Not All Tokens Are Worth Caching: Learning Semantic-Aware Eviction for LLM Prefix Caches](https://mlsanity.com/paper/2605.18825/) (2026)
- [KVBuffer: IO-aware Serving for Linear Attention](https://mlsanity.com/paper/2605.19049/) (2026)
- [Bole: Efficient Tree Speculation for Hybrid-Attention Language Models](https://mlsanity.com/paper/2608.01651/) (2026)
- [LinearKV: One Cached State Suffices for Position-Independent Caching in Hybrid LLMs](https://mlsanity.com/paper/2608.11231/) (2026)
- [ScaleSim: Serving Large-Scale Multi-Agent Simulation with Invocation Distance-Based Memory Management](https://mlsanity.com/paper/2601.21473/) (2026)
- [DASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving](https://mlsanity.com/paper/2608.30386/) (2026)
- [ReplaySSM: Cache SSM Inputs, Not State](https://mlsanity.com/paper/2610.13778/) (2026)
- [Tail-Replay: Escaping the Curse of Linear Attention in Prefix Caching for Hybrid LLMs](https://mlsanity.com/paper/2608.30310/) (2026)
- [Intra-Prompt Parallel Decoding for Common-Context Question Answering](https://mlsanity.com/paper/2609.05707/) (2026)
- [Recency/Frequency Adaptive KV Caching for Large Language Model Serving](https://mlsanity.com/paper/2606.21238/) (2026)
- [ASAP: A Disaggregated and Asynchronous Inference System for MoE Prefill](https://mlsanity.com/paper/2606.22541/) (2026)
- [PEEK: Predictive Queue-Informed KV Cache Management for LLM Serving](https://mlsanity.com/paper/2607.02525/) (2026)
- [Shared KV Caching for Replicated 27B Inference: Correctness Failures and Performance Boundaries](https://mlsanity.com/paper/2609.15021/) (2026)
- [Validating Hybrid-State Cache Recovery for GLM-5.3-Flash with vLLM and LMCache](https://mlsanity.com/paper/2609.15030/) (2026)
- [LumoTree: Path-Parallel Speculative Verification for Hybrid Language Models](https://mlsanity.com/paper/2609.23900/) (2026)
- [LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay](https://mlsanity.com/paper/2609.25053/) (2026)
- [Just Let Linear States Forget the Distant Past: Prefix Caching via Suffix Replay for Hybrid LLMs](https://mlsanity.com/paper/2609.33477/) (2026)
- [EfficientAgent: What Makes KV Cache Offloading Work for Concurrent Agents?](https://mlsanity.com/paper/2609.33762/) (2026)
- [WavePP: High-Throughput Pipeline Parallel LLM Prefill under Prefix Reuse](https://mlsanity.com/paper/2609.35263/) (2026)
- [LeapQuant: Efficient Linear Attention with Accurate Recurrent State Quantization](https://mlsanity.com/paper/2609.38166/) (2026)

## Abstract (arXiv)

Hybrid models that combine the language modeling capabilities of Attention layers with the efficiency of Recurrent layers (e.g., State Space Models) have gained traction in practically supporting long contexts in Large Language Model serving. Yet, the unique properties of these models complicate the usage of complementary efficiency optimizations such as prefix caching that skip redundant computations across requests. Most notably, their use of in-place state updates for recurrent layers precludes rolling back cache entries for partial sequence overlaps, and instead mandates only exact-match cache hits; the effect is a deluge of (large) cache entries per sequence, most of which yield minimal reuse opportunities. We present Marconi, the first system that supports efficient prefix caching with Hybrid LLMs. Key to Marconi are its novel admission and eviction policies that more judiciously assess potential cache entries based not only on recency, but also on (1) forecasts of their reuse likelihood across a taxonomy of different hit scenarios, and (2) the compute savings that hits deliver relative to memory footprints. Across diverse workloads and Hybrid models, Marconi achieves up to 34.4× higher token hit rates (71.1% or 617 ms lower TTFT) compared to state-of-the-art prefix caching systems.