mlsanityData to 7 Oct 2026

GTM: Simulating the World of Tools for AI Agents

arXiv 2512.04535 ↗First posted 4 Dec 2025cs.AIcited by 12

On the map: LLM agents and tool use → Reinforcement learning for LLM agents

Abstract (arXiv)

The integration of external tools is pivotal for empowering Large Language Model (LLM) agents with real-world capabilities. However, training these agents through direct, continuous interaction with diverse tools is often prohibitively expensive, slow, and introduces additional development and maintenance overhead. To address this challenge, we introduce the Generalist Tool Model (GTM), a 1.5-billion-parameter model that learns to act as a universal tool simulator. With only prompt-level configuration, GTM accesses tool functionalities along with input arguments and generates outputs that faithfully mimic real tool execution, providing a fast and cost-effective solution that eliminates development overhead. To build GTM, we propose the Context-Aware Response Generation (CARG) pipeline, which synthesizes comprehensive training data covering over 20,000 tools across 300 domains including physics, medicine, robotics, and finance. Through this pipeline, GTM learns to produce not only syntactically correct outputs but also logically coherent and contextually appropriate responses. Experiments demonstrate that GTM produces high-quality outputs with strong consistency and reliability. Besides when used in real reinforcement learning scenarios for agent training, GTM exhibits significantly faster simulation speed compared to real tools while maintaining comparable output quality, along with remarkable generalization and domain adaptability. Our results establish GTM as a foundational component for developing future AI agents, enabling efficient and scalable training of tool-augmented systems.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use. Builds on 38 of its 42 references that are in the map. First posted on arXiv 4 Dec 2025.

Builds on

38 of 42 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Agent Lightning: Train ANY AI Agents with Reinforcement Learning5 Aug
  2. 2025Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning28 Apr
  3. 2025ToolRL: Reward is All Tool Learning Needs16 Apr
  4. 2025ReTool: Reinforcement Learning for Strategic Tool Use in LLMs15 Apr
  5. 2025Reinforcement Learning for Long-Horizon Interactive LLM Agents3 Feb

Other references · 33

  1. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  2. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  3. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
  4. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,371
  5. 2024Executable Code Actions Elicit Better LLM Agentscited by 537
  6. 2023ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Casescited by 383
  7. 2025WebThinker: Empowering Large Reasoning Models with Deep Research Capabilitycited by 320
  8. 2023ToolQA: A Dataset for LLM Question Answering with External Toolscited by 273
  9. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 268
  10. 2025Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learningcited by 260
  11. 2025Inference-Time Scaling for Generalist Reward Modelingcited by 233
  12. 2024ToolACE: Winning the Points of LLM Function Callingcited by 204
21 more
  1. 2025ZeroSearch: Incentivize the Search Capability of LLMs without Searchingcited by 203
  2. 2024APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasetscited by 188
  3. 2024The Landscape of Emerging AI Agent Architectures for Reasoning, Planning, and Tool Calling: A Surveycited by 164
  4. 2025WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learningcited by 162
  5. 2024Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methodscited by 114
  6. 2024Reinforcement Learning Enhanced LLMs: A Surveycited by 69
  7. 2024Seal-Tools: Self-Instruct Tool Learning Dataset for Agent Tuning and Detailed Benchmarkcited by 62
  8. 2024Tool Learning in the Wild: Empowering Language Models as Automatic Tool Agentscited by 56
  9. 2025MedAgentBench: A Realistic Virtual EHR Environment to Benchmark Medical LLM Agentscited by 56
  10. 2025Towards General Agentic Intelligence via Environment Scalingcited by 51
  11. 2024API-BLEND: A Comprehensive Corpora for Training and Benchmarking API LLMscited by 39
  12. 2024ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarioscited by 35
  13. 2025Generalist Reward Models: Found Inside Large Language Modelscited by 34
  14. 2025R1-Searcher++: Incentivizing the Dynamic Knowledge Acquisition of LLMs via Reinforcement Learningcited by 30
  15. 2024ToolFlow: Boosting LLM Tool-Calling Through Natural and Coherent Dialogue Synthesiscited by 27
  16. 2024Harnessing LLMs for API Interactions: A Framework for Classification and Synthetic Data Generationcited by 17
  17. 2025T1: A Tool-Oriented Conversational Dataset for Multi-Turn Agentic Planningcited by 14
  18. 2025LMDeploy Accelerates Mixed-Precision LLM Inference with TurboMindcited by 10
  19. 2025Advancing and Benchmarking Personalized Tool Invocation for LLMscited by 7
  20. 2024MetaTool: Facilitating Large Language Models to Master Tools with Meta-task Augmentationcited by 6
  21. 2025Help or Hurdle? Rethinking Model Context Protocol-Augmented Large Language Modelscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 1; 2026: 11.

12 papers.

  1. 2025AutoForge: Automated Environment Synthesis for Agentic Reinforcement Learningcited by 14
  2. 2026CM2: Reinforcement Learning with Checklist Rewards for Multi-Turn and Multi-Step Agentic Tool Usecited by 13
  3. 2026Agentic World Modeling: Foundations, Capabilities, Laws, and Beyondcited by 10
  4. 2026Budget-Constrained Agentic Large Language Models: Intention-Based Planning for Costly Tool Usecited by 8
  5. 2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
  6. 2026Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Modelcited by 1
  7. 2026Do Enterprise Systems Need Learned World Models? The Importance of Context to Infer Dynamicscited by 1
  8. 2026RIFT-Bench: Dynamic Red-teaming For Agentic AI Systemscited by 0
  9. 2026Quo Vadis, World Modeling?cited by 0
  10. 2026On Understanding, Identifying, and Mitigating Vulnerabilities in Agentic Large Language Modelscited by 0
  11. 2026Agent Seer: Synthesizing Scenarios from Specification Understandingcited by 0
  12. 2026Continual Enterprise World Model Discovery in Dynamic Systemscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Tool-R1: Sample-Efficient Reinforcement Learning for Agentic Tool Use
  2. 2025Generalizable End-to-End Tool-Use RL with Synthetic CodeGym
  3. 2024AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
  4. 2026Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
  5. 2024GTA: A Benchmark for General Tool Agents
  6. 2024xLAM: A Family of Large Action Models to Empower AI Agent Systems
  7. 2024Learning to Use Tools via Cooperative and Interactive Agents
  8. 2025ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering