Data to 7 Oct 2026

TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments

arXiv 2510.01179 ↗First posted 1 Oct 2025cs.LG cs.AI cs.CLcited by 61

On the map: LLM agents and tool use → Tool learning and function calling

Abstract (arXiv)

Large Language Model (LLM) agents are rapidly emerging as powerful systems for automating tasks across domains. Yet progress in the open-source community is constrained by the lack of high quality permissively licensed tool-agentic training data. Existing datasets are often limited in diversity, realism, and complexity, particularly regarding multi-tool and multi-turn interactions. To address this gap, we introduce Toucan, the largest publicly available tool-agentic dataset to date, containing 1.5 million trajectories synthesized from nearly 500 real-world Model Context Protocols (MCPs). Unlike prior work, Toucan leverages authentic MCP environments to generate diverse, realistic, and challenging tasks with trajectories involving real tool execution. Our pipeline first produces a broad spectrum of tool-use queries using five distinct models, applies model-based quality filtering, and then generates agentic trajectories with three teacher models using two agentic frameworks. Rigorous rule-based and model-based validation ensures high-quality outputs. We also introduce three extension mechanisms to further diversify tasks and simulate multi-turn conversations. Models fine-tuned on Toucan outperform larger closed-source counterparts on the BFCL V3 benchmark and push the Pareto frontier forward on MCP-Universe Bench.

Key facts. Cited by 61 papers in the map (data through 7 Oct 2026). Placed in Tool learning and function calling, one of 16 directions in LLM agents and tool use. Builds on 19 of its 36 references that are in the map. First posted on arXiv 1 Oct 2025.

Builds on

19 of 36 references.

Closest ancestors · 10

Same direction, newest first

Other references · 9

2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
2024τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domainscited by 1,280
2025τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environmentcited by 558
2025APIGen-MT: Agentic Pipeline for Multi-Turn Data Generation via Simulated Agent-Human Interplaycited by 158
2025From LLM Reasoning to Autonomous AI Agents: A Comprehensive Reviewcited by 146
2025KodCode: A Diverse, Challenging, and Verifiable Synthetic Dataset for Codingcited by 115
2025MCPWorld: A Unified Benchmarking Testbed for API, GUI, and Hybrid Computer Use Agentscited by 37
2024API Pack: A Massive Multi-Programming Language Dataset for API Call Generationcited by 5
2025A Scalable Approach to Clustering Embedding Projectionscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 10; 2026: 51.

61 papers.

Sort:

By area:

2025MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scalingcited by 72
2025Simulating Environments with Reasoning Models for Agent Trainingcited by 52
2026Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligencecited by 35
2026Nemotron 3 Super: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoningcited by 33
2025ToolMind Technical Report: A Large-Scale, Reasoning-Enhanced Tool-Use Datasetcited by 27
2025LoopTool: Closing the Data-Training Loop for Robust LLM Tool Callscited by 16
2025AutoForge: Automated Environment Synthesis for Agentic Reinforcement Learningcited by 14
2026Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Textcited by 13
2026EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RLcited by 11
2025Youtu-LLM: Unlocking the Native Agentic Potential for Lightweight Large Language Modelscited by 8
2026CacheWise: Understanding Workloads and Optimizing KVCache Management for Efficiently Serving LLM Coding Agentscited by 8
2026ASTRA: Automated Synthesis of agentic Trajectories and Reinforcement Arenascited by 6

38 more

2026PhGPO: Pheromone-Guided Policy Optimization for Long-Horizon Tool Planningcited by 6
2025Delegated Authorization for Agents Constrained to Semantic Task-to-Scope Matchingcited by 5
2026LOGIGEN: Logic-Driven Generation of Verifiable Agentic Taskscited by 5
2026EigenData: A Self-Evolving Multi-Agent Platform for Function-Calling Data Synthesis, Auditing, and Repaircited by 5
2026Controllable and Verifiable Tool-Use Data Synthesis for Agentic Reinforcement Learningcited by 5
2026Learning to Rewrite Tool Descriptions for Reliable LLM-Agent Tool Usecited by 4
2026Tool-Genesis: A Task-Driven Tool Creation Benchmark for Self-Evolving Language Agentcited by 4
2026SING: Synthetic Intention Graph for Scalable Active Tool Discovery in LLM Agentscited by 4
2025K2-V2: A 360-Open, Reasoning-Enhanced LLMcited by 3
2026Simulating Complex Multi-Turn Tool Calling Interactions in Stateless Execution Environmentscited by 3
2026UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agentscited by 3
2026ToolVerse: Unlocking Massive Environments and Long-Horizon Tasks for Agentic Reinforcement Learningcited by 3
2026ReCache: Efficient KV Cache Reuse and Compression for Tool-Augmented LLM Agentscited by 3
2026Typhoon-S: Minimal Open Post-Training for Sovereign Large Language Modelscited by 2
2026ARTIS: Agentic Risk-Aware Test-Time Scaling via Iterative Simulationcited by 2
2026Firefly: Illuminating Large-Scale Verified Tool-Call Data Generation from Real APIscited by 2
2026RODS: Reward-Driven Online Data Synthesis for Multi-Turn Tool-Use Agentscited by 2
2026EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learningcited by 2
2026What Makes Good Agentic Data? An ACE Lens on Data Generation for LLM Agentscited by 2
2026AgentBrew: Offline Tool-Use Agent Learning from Raw Real-World Trajectoriescited by 2
2025AMAP Agentic Planning Technical Reportcited by 1
2026Dicta-LM 3.0: Advancing The Frontier of Hebrew Sovereign LLMscited by 1
2026Fine-Tuning Small Reasoning Models for Quantum Field Theorycited by 1
2026When the Strongest Teacher Is Not the Best Teacher: Student-Centric Answer Selectioncited by 1
2026WRIT: Write-Read Intensive Trajectory Synthesis for Multi-Turn User-Facing Agentscited by 1
2026ISE: An Execution-Grounded Recipe for Multi-Turn OS-Agent Trajectoriescited by 1
2026Execution-First Synthetic Tool-Use Trace Generation for LLM Agentscited by 1
2026OctoLong: Mid-Training On Cross-Repository Code Contexts Enhances Long-Context Modelingcited by 1
2026Learning Generalizable Behaviors for Terminal Agentscited by 1
2026EnvCraft: Synthesizing Executable Environments in Agentic RL for Claw-like Agentcited by 1
2026NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harnesscited by 1
2025HumanMCP: A Human-Like Query Dataset for Evaluating MCP Tool Retrieval Performancecited by 0
2026NGDBench: Towards Neural Graph Data Managementcited by 0
2026A Matter of TASTE: Improving Coverage and Difficulty of Agent Benchmarkscited by 0
2026Towards Direct Latent-Space Synthesis for Parallel Branches in LLM-Agent Workflowscited by 0
2026From Knowing to Acting: Benchmarking Self-Awareness Capability of LLM Agentscited by 0
2026The Open Source Economic Index of AI Adoption and Capabilitycited by 0
2026UMoE:Unlocking Every Expert in Domain-Specific Trainingcited by 0

50 most cited of 61 All 61 (JSON)

Similar, not cited

Close in wording, no citation either way