# AgentStop: Terminating Local AI Agents Early to Save Energy in Consumer Devices

arXiv: https://arxiv.org/abs/2605.15206 · HTML: https://mlsanity.com/paper/2605.15206/

## Key facts

- Cited by 6 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 16 of its 37 references that are in the map.
- First posted on arXiv 1 May 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, 16 of 16)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [XGBoost: A Scalable Tree Boosting System](https://mlsanity.com/paper/1603.02754/) (2016)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [BERT Loses Patience: Fast and Robust Inference with Early Exit](https://mlsanity.com/paper/2006.04152/) (2020)
- [Small Language Models are the Future of Agentic AI](https://mlsanity.com/paper/2506.02153/) (2025)
- [When Do Neural Nets Outperform Boosted Trees on Tabular Data?](https://mlsanity.com/paper/2305.02997/) (2023)
- [Fact, Fetch, and Reason: A Unified Evaluation of Retrieval-Augmented Generation](https://mlsanity.com/paper/2409.12941/) (2024)
- [Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning](https://mlsanity.com/paper/2310.03094/) (2023)
- [Language Model Cascades: Token-level uncertainty and beyond](https://mlsanity.com/paper/2404.10136/) (2024)
- [The Cost of Dynamic Reasoning: Demystifying AI Agents and Test-Time Scaling from an AI Infrastructure Perspective](https://mlsanity.com/paper/2506.04301/) (2025)
- [MELTing point: Mobile Evaluation of Language Transformers](https://mlsanity.com/paper/2403.12844/) (2024)
- [Measuring the environmental impact of delivering AI at Google Scale](https://mlsanity.com/paper/2508.15734/) (2025)
- [A Simple Hash-Based Early Exiting Approach For Language Understanding and Generation](https://mlsanity.com/paper/2203.01670/) (2022)
- [Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments](https://mlsanity.com/paper/2505.17616/) (2025)
- [PalmBench: A Comprehensive Benchmark of Compressed Large Language Models on Mobile Platforms](https://mlsanity.com/paper/2410.05315/) (2024)
- [Typical Phone Use Habits: Intense Use Does Not Predict Negative Well-Being](https://mlsanity.com/paper/1807.02472/) (2018)

## Built on by (most cited first, 6 of 6)

- [Doomed from the Start: Early Abort of LLM Agent Episodes via a Recall-Controlled Probe Cascade](https://mlsanity.com/paper/2607.06503/) (2026)
- [EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction](https://mlsanity.com/paper/2609.02783/) (2026)
- [Before the Rollout Ends: Early Terminal Reward Prediction for Long-horizon Coding Agents](https://mlsanity.com/paper/2609.31995/) (2026)
- [Long-Horizon Scaling: How Model Capabilities Shape the Returns to Computation](https://mlsanity.com/paper/2609.35236/) (2026)
- [What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agents](https://mlsanity.com/paper/2609.39989/) (2026)
- [Disentangling Task Difficulty from Run-Level Failure in Agent Failure Prediction](https://mlsanity.com/paper/2610.05572/) (2026)

## Abstract (arXiv)

Autonomous agents powered by large language models (LLMs) are increasingly used to automate complex, multi-step tasks such as coding or web-based question answering. While remote, cloud-based agents offer scalability and ease of deployment, they raise privacy concerns, depend on network connectivity, and incur recurring API costs. Deploying agents locally on user devices mitigates these issues by preserving data privacy and eliminating usage-based fees. However, agentic workflows are far more resource-intensive than typical LLM interactions. Iterative reasoning, tool use, and failure retries substantially increase token consumption, often expending significant compute without successfully completing tasks. In this work, we investigate the time, token, and energy overhead of locally deployed LLM-based agents on consumer hardware. Our measurements show that agentic execution increases GPU power draw, temperature, and battery drain compared to single-inference workloads. To address this inefficiency, we introduce AgentStop, a lightweight efficiency supervisor that predicts and preemptively terminates trajectories unlikely to succeed. Leveraging low-cost execution signals, such as token-level log probabilities, AgentStop can reduce wasted energy by 15-20% with minimal impact on task performance (<5% utility drop) for challenging web-based question answering and coding benchmarks. These findings position predictive early termination as a practical mechanism for enabling sustainable, privacy-preserving LLM agents on user devices. Our project code and data are available at https://github.com/brave-experiments/AgentStop.