AgentVista: Evaluating Multimodal Agents in Ultra-Challenging Realistic Visual Scenarios
This work addresses the critical need for a more realistic and challenging benchmark for generalist multimodal agents, particularly for developers aiming to build agents capable of complex, real-world problem-solving.
The authors introduce AgentVista, a new benchmark to evaluate multimodal agents in complex, realistic visual scenarios that require multi-step workflows and long-horizon tool use. State-of-the-art models, including Gemini-3-Pro with tools, achieved only 27.3% overall accuracy, highlighting significant limitations in current agent capabilities.
Real-world multimodal agents solve multi-step workflows grounded in visual evidence. For example, an agent can troubleshoot a device by linking a wiring photo to a schematic and validating the fix with online documentation, or plan a trip by interpreting a transit map and checking schedules under routing constraints. However, existing multimodal benchmarks mainly evaluate single-turn visual reasoning or specific tool skills, and they do not fully capture the realism, visual subtlety, and long-horizon tool use that practical agents require. We introduce AgentVista, a benchmark for generalist multimodal agents that spans 25 sub-domains across 7 categories, pairing realistic and detail-rich visual scenarios with natural hybrid tool use. Tasks require long-horizon tool interactions across modalities, including web search, image search, page navigation, and code-based operations for both image processing and general programming. Comprehensive evaluation of state-of-the-art models exposes significant gaps in their ability to carry out long-horizon multimodal tool use. Even the best model in our evaluation, Gemini-3-Pro with tools, achieves only 27.3% overall accuracy, and hard instances can require more than 25 tool-calling turns. We expect AgentVista to accelerate the development of more capable and reliable multimodal agents for realistic and ultra-challenging problem solving.