HASHIRU: Hierarchical Agent System for Hybrid Intelligent Resource Utilization
On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks
Abstract (arXiv)
Rapid Large Language Model (LLM) advancements are fueling autonomous Multi-Agent System (MAS) development. However, current frameworks often lack flexibility, resource awareness, model diversity, and autonomous tool creation. This paper introduces HASHIRU (Hierarchical Agent System for Hybrid Intelligent Resource Utilization), a novel MAS framework enhancing flexibility, resource efficiency, and adaptability. HASHIRU features a "CEO" agent dynamically managing specialized "employee" agents, instantiated based on task needs and resource constraints (cost, memory). Its hybrid intelligence prioritizes smaller, local LLMs (via Ollama) while flexibly using external APIs and larger models when necessary. An economic model with hiring/firing costs promotes team stability and efficient resource allocation. The system also includes autonomous API tool creation and a memory function. Evaluations on tasks like academic paper review (58% success), safety assessments (100% on a JailbreakBench subset), and complex reasoning (outperforming Gemini 2.0 Flash on GSM8K: 96% vs. 61%; JEEBench: 80% vs. 68.3%; SVAMP: 92% vs. 84%) demonstrate HASHIRU's capabilities. Case studies illustrate its self-improvement via autonomous cost model generation, tool integration, and budget management. HASHIRU offers a promising approach for more robust, efficient, and adaptable MAS through dynamic hierarchical control, resource-aware hybrid intelligence, and autonomous functional extension. Source code and benchmarks are available at https://github.com/HASHIRU-AI/HASHIRU and https://github.com/HASHIRU-AI/HASHIRUBench respectively, and a live demo is available at https://hashiruagentx-hashiruai.hf.space upon request.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 31 of its 64 references that are in the map. First posted on arXiv 1 Jun 2025.
Builds on
31 of 64 references.
Closest ancestors · 1
Same direction, newest first
Other references · 30
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2022Training language models to follow instructions with human feedbackcited by 15,971
- 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
- 2021Training Verifiers to Solve Math Word Problemscited by 10,109
- 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
- 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
- 2021Evaluating Large Language Models Trained on Codecited by 8,320
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
- 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,276
- 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
18 more
- 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
- 2019Fine-Tuning Language Models from Human Preferencescited by 2,150
- 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,988
- 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
- 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,463
- 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
- 2021Are NLP Models really able to Solve Simple Math Word Problems?cited by 878
- 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
- 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Modelscited by 536
- 2024Benchmarking Complex Instruction-Following with Multiple Constraints Compositioncited by 152
- 2023Emergent autonomous scientific research capabilities of large language modelscited by 132
- 2023Have LLMs Advanced Enough? A Challenging Problem Solving Benchmark For Large Language Modelscited by 77
- 2018A Systematic Classification of Knowledge, Reasoning, and Context within the ARC Datasetcited by 55
- 2024Self-evolving Agents with reflective and memory-augmented abilitiescited by 52
- 2024Calibration and Correctness of Language Models for Codecited by 49
- 2024Hermes 3 Technical Reportcited by 34
- 2024QA-Calibration of Language Model Confidence Scorescited by 13
- 2025CoDocBench: A Dataset for Code-Documentation Alignment in Software Maintenancecited by 5
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025COALESCE: Economic and Security Dynamics of Skill-Based Task Outsourcing Among Team of Autonomous LLM Agents
- 2025Nexus: A Lightweight and Scalable Multi-Agent Framework for Complex Tasks Automation
- 2025OmniNova:A General Multimodal Agent Framework
- 2024Understanding the Weakness of Large Language Model Agents within a Complex Android Environment
- 2025Youtu-Agent: Scaling Agent Productivity with Automated Generation and Hybrid Policy Optimization
- 2026DEALS: Decentralized Expertise-Aware Load Serving for Multi-Agent LLM Systems
- 2026LLM-Enabled Multi-Agent Systems: Empirical Evaluation and Insights into Emerging Design Patterns & Paradigms
- 2026ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents