VTC-Bench: Evaluating Agentic Multimodal Models via Compositional Visual Tool Chaining

TL;DR

VTC-Bench is a comprehensive benchmark designed to evaluate multimodal models' ability to compose and execute diverse visual tools in complex, multi-step tasks, revealing current limitations and guiding future improvements.

Contribution

The paper introduces VTC-Bench, a new benchmark with 32 visual operations and 680 problems to assess tool-use proficiency and multi-tool composition in multimodal large language models.

Findings

Current models struggle with diverse tool-sets and unseen operations.

Models have difficulty formulating efficient multi-step execution plans.

Leading model achieves only 51% on the benchmark.

Abstract

Recent advancements extend Multimodal Large Language Models (MLLMs) beyond standard visual question answering to utilizing external tools for advanced visual tasks. Despite this progress, precisely executing and effectively composing diverse tools for complex tasks remain persistent bottleneck. Constrained by sparse tool-sets and simple tool-use trajectories, existing benchmarks fail to capture complex and diverse tool interactions, falling short in evaluating model performance under practical, real-world conditions. To bridge this gap, we introduce VisualToolChain-Bench(VTC-Bench), a comprehensive benchmark designed to evaluate tool-use proficiency in MLLMs. To align with realistic computer vision pipelines, our framework features 32 diverse OpenCV-based visual operations. This rich tool-set enables extensive combinations, allowing VTC-Bench to rigorously assess multi-tool composition and long-horizon, multi-step plan execution. For precise evaluation, we provide 680 curated problems structured across a nine-category cognitive hierarchy, each with ground-truth execution trajectories. Extensive experiments on 19 leading MLLMs reveal critical limitations in current models' visual agentic capabilities. Specifically, models struggle to adapt to diverse tool-sets and generalize to unseen operations, with the leading model Gemini-3.0-Pro only achieving 51% on our benchmark. Furthermore, multi-tool composition remains a persistent challenge. When facing complex tasks, models struggle to formulate efficient execution plans, relying heavily on a narrow, suboptimal subset of familiar functions rather than selecting the optimal tools. By identifying these fundamental challenges, VTC-Bench establishes a rigorous baseline to guide the development of more generalized visual agentic models.