Long Live the Librarian! A Persistent Search Sub-Agent for Energy-Efficient Multi-Agent Software Engineering Systems
cs.MA, cs.CL
Submitted: 2026-05-27
Updated: 2026-08-31
Comments: 2026 EMNLP Main Conference
License: http://creativecommons.org/licenses/by/4.0/
The gist: Multi-agent systems (MAS) have substantially advanced autonomous software engineering (SWE), but their growing inference energy demands raise sustainability concerns.
Terminology
Abstract
Multi-agent systems (MAS) have substantially advanced autonomous software engineering (SWE), but their growing inference energy demands raise sustainability concerns. In this paper, we demonstrate that this cost is concentrated in an overlooked source: redundant output tokens generated across agents. Two empirical findings ground this claim. First, our per-token energy attribution for MAS reveals a sharp asymmetry: an output token consumes 30 to 1,000 times more energy than an input or cached token. Second, MAS inflate per-episode output because agents repeatedly re-explore overlapping repository regions. To address this inefficiency, we propose Librarian, a persistent search sub-agent that tracks repository-search history and suppresses redundant exploration actions across agents. By returning short references to file regions instead of full file excerpts, Librarian further reduces output-token volume. On SWE-Bench Verified and SWE-Bench-Live, Librarian reduces per-episode GPU energy consumption of existing multi-agent SWE systems by 11 to 30% while preserving task performance. Our code is available at https://github.com/ml-postech/Librarian.
Sources
- Brevity Constraints Reverse Performance Hierarchies in Language Models
- CodeXEmbed: A Generalist Embedding Model Family for Multiligual and Multi-task Code Retrieval
- RepoFusion: Training Code Models to Understand Your Repository
- MemGPT: Towards LLMs as Operating Systems
- SWEnergy: An Empirical Study on Energy Efficiency in Agentic Issue Resolution Frameworks with SLMs
- HyperAgent: Generalist Software Engineering Agents to Solve Coding Tasks at Scale
Related papers
- Highway Congestion Reduction through Reinforcement Learning Based Eulerian Headway Control
- Temporal Fair Division in Multi-Agent Systems: From Precise Alternation Metrics to Scalable Coordination Proxies
- Auditable and Deployable Multi-Agent Motion Planning via Posetal Games from Legal Guidelines
- ORACLE-SWE: Quantifying the Contribution of Critical Information Signals on SWE Agents
- You Only Align Once: Propagating Cooperative Behaviors in Multi-Agent Systems through Seed Agents
- Deny Without Disabling: Authorization-Paired Evaluation and Control for Multi-Agent Systems