BFCL v-3

The 'BFCL-v3' dataset/benchmark is used to evaluate the performance of Large Language Models (LLMs) in executing complex, multi-step tasks by providing a structured set of data that reflects the model's capabilities and weaknesses.

Papers using BFCL v-3 (4)

Can a Single Model Master Both Multi-turn Conversations and Tool Use? CALM: A Unified Conversational Agentic Language Model2025TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments2025DaMo: Data Mixing Optimizer in Fine-tuning Multimodal LLMs for Mobile Phone Agents2025LoopTool: Closing the Data-Training Loop for Robust LLM Tool Calls2025