# Don't Break the Cache: An Evaluation of Prompt Caching for Long-Horizon Agentic Tasks

arXiv: https://arxiv.org/abs/2601.06007 · HTML: https://mlsanity.com/paper/2601.06007/

## Key facts

- Cited by 19 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.
- Builds on 29 of its 55 references that are in the map.
- First posted on arXiv 9 Jan 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 29 of 29)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents](https://mlsanity.com/paper/2504.12516/) (2025)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [Efficiently Scaling Transformer Inference](https://mlsanity.com/paper/2211.05102/) (2022)
- [Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs](https://mlsanity.com/paper/2310.01801/) (2023)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?](https://mlsanity.com/paper/2403.07718/) (2024)
- [Prompt Cache: Modular Attention Reuse for Low-Latency Inference](https://mlsanity.com/paper/2311.04934/) (2023)
- [DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents](https://mlsanity.com/paper/2506.11763/) (2025)
- [CacheBlend: Fast Large Language Model Serving for RAG with Cached Knowledge Fusion](https://mlsanity.com/paper/2405.16444/) (2024)
- [A Survey on Efficient Inference for Large Language Models](https://mlsanity.com/paper/2404.14294/) (2024)
- [CacheGen: KV Cache Compression and Streaming for Fast Large Language Model Serving](https://mlsanity.com/paper/2310.07240/) (2023)
- [LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference](https://mlsanity.com/paper/2510.09665/) (2025)
- [Keep the Cost Down: A Review on Methods to Optimize LLM' s KV-Cache Consumption](https://mlsanity.com/paper/2407.18003/) (2024)
- [Hydragen: High-Throughput LLM Inference with Shared Prefixes](https://mlsanity.com/paper/2402.05099/) (2024)
- [Seal-Tools: Self-Instruct Tool Learning Dataset for Agent Tuning and Detailed Benchmark](https://mlsanity.com/paper/2405.08355/) (2024)
- [Layer-Condensed KV Cache for Efficient Inference of Large Language Models](https://mlsanity.com/paper/2405.10637/) (2024)
- [Evaluating LLM-based Agents for Multi-Turn Conversations: A Survey](https://mlsanity.com/paper/2503.22458/) (2025)
- [Re-Invoke: Tool Invocation Rewriting for Zero-Shot Tool Retrieval](https://mlsanity.com/paper/2408.01875/) (2024)
- [ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool Retrieval](https://mlsanity.com/paper/2403.06551/) (2024)
- [ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks](https://mlsanity.com/paper/2508.15804/) (2025)
- [ScaleMCP: Dynamic and Auto-Synchronizing Model Context Protocol Tools for LLM Agents](https://mlsanity.com/paper/2505.06416/) (2025)
- [Compute Or Load KV Cache? Why Not Both?](https://mlsanity.com/paper/2410.03065/) (2024)
- [Toolshed: Scale Tool-Equipped Agents with Advanced RAG-Tool Fusion and Tool Knowledge Bases](https://mlsanity.com/paper/2410.14594/) (2024)
- [Auditing Prompt Caching in Language Model APIs](https://mlsanity.com/paper/2502.07776/) (2025)
- [Do Large Language Models Need a Content Delivery Network?](https://mlsanity.com/paper/2409.13761/) (2024)
- [ContextCache: Context-Aware Semantic Cache for Multi-Turn Queries in Large Language Models](https://mlsanity.com/paper/2506.22791/) (2025)

## Built on by (most cited first, 19 of 19)

- [Self-GC: Self-Governing Context for Long-Horizon LLM Agents](https://mlsanity.com/paper/2607.00692/) (2026)
- [Recursive Agent Harnesses](https://mlsanity.com/paper/2606.13643/) (2026)
- [Beyond Rows to Reasoning: Agentic Retrieval for Multimodal Spreadsheet Understanding and Editing](https://mlsanity.com/paper/2603.06503/) (2026)
- [Ask Early, Ask Late, Ask Right: When Does Clarification Timing Matter for Long-Horizon Agents?](https://mlsanity.com/paper/2605.07937/) (2026)
- [AgenticSTS: A Bounded-Memory Testbed for Long-Horizon LLM Agents](https://mlsanity.com/paper/2607.02255/) (2026)
- [The Workload-Router-Pool Architecture for LLM Inference Optimization: A Vision Paper from the vLLM Semantic Router Project](https://mlsanity.com/paper/2603.21354/) (2026)
- [Evaluating Memory Condensation Strategies for Coding Agents in Data-Driven Scientific Discovery](https://mlsanity.com/paper/2605.18854/) (2026)
- [Cache-Aware Prompt Compression:A Two-Tier Cost Model for LLM API Caching](https://mlsanity.com/paper/2607.15516/) (2026)
- [ARES: Adaptive Reasoning-Effort Steering for PPA- and Cost-Aware RTL Optimization with LLM Agents](https://mlsanity.com/paper/2607.27879/) (2026)
- [Control the Harness, Control the Cost: Routing and Governing AI Coding Agents in the Enterprise](https://mlsanity.com/paper/2609.28919/) (2026)
- [Agentic Jackal: Live Execution and Semantic Value Grounding for Text-to-JQL](https://mlsanity.com/paper/2604.09470/) (2026)
- [Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution](https://mlsanity.com/paper/2607.08700/) (2026)
- [ACEM: A Cost Estimation Model for Agentic Software Engineering](https://mlsanity.com/paper/2608.02582/) (2026)
- [Parsing the Stream: A Live Trace Model for Long-Horizon Agents and Their Observers](https://mlsanity.com/paper/2609.01466/) (2026)
- [Same Request, Different Answer: Quantization Amplifies Cache-Induced Divergence in LLM Serving](https://mlsanity.com/paper/2609.04748/) (2026)
- [Protocol Compression Changes Which Party Pays: Bilateral Cost in Cross-Organization LLM Agent Communication](https://mlsanity.com/paper/2609.06129/) (2026)
- [ActKV: Efficient LLM Agents through Action-Guided KV Cache Management](https://mlsanity.com/paper/2609.31395/) (2026)
- [Positions Are Not Facts: The Mismatch Between KV Caches and Memory](https://mlsanity.com/paper/2609.33759/) (2026)
- [CADOC: Cache-Aware Dynamic Object Context for Long-Horizon Agents](https://mlsanity.com/paper/2609.37012/) (2026)

## Abstract (arXiv)

Recent advancements in Large Language Model (LLM) agents have enabled complex multi-turn agentic tasks requiring extensive tool calling, where conversations can span dozens of API calls with increasingly large context windows. However, although major LLM providers offer prompt caching to reduce cost and latency, its benefits for agentic workloads remain underexplored in the research literature. To our knowledge, no prior work quantifies these cost savings or compares caching strategies for multi-turn agentic tasks. We present a comprehensive evaluation of prompt caching across three major LLM providers (OpenAI, Anthropic, and Google) and compare three caching strategies, including full context caching, system prompt only caching, and caching that excludes dynamic tool results. We evaluate on DeepResearch Bench, a multi-turn agentic benchmark where agents autonomously execute real-world web search tool calls to answer complex research questions, measuring both API cost and time to first token (TTFT) across over 500 agent sessions with 10,000-token system prompts. Our results demonstrate that prompt caching reduces API costs by 41-80% and improves time to first token by 13-31% across providers. We find that strategic prompt cache block control, such as placing dynamic content at the end of the system prompt, avoiding dynamic traditional function calling, and excluding dynamic tool results, provides more consistent benefits than naive full-context caching, which can paradoxically increase latency. An ablation study across prompt sizes (500-50,000 tokens) and tool call counts (3-50) demonstrates universal linear cost and TTFT benefits, after the provider caching token minimum, and reveal provider-specific strategy discrepancies across variants. We provide nuanced discussion and guidance for implementing prompt caching in production agentic systems.