JarvisGUI Benchmark

What is JarvisGUI?

Proposed benchmark for cross-device GUI agent workflows across Android, Windows, and Ubuntu, using input-output transformations and dynamic task composition.

Released

2026-09-09

Evaluates

Agents & Tool Use, Mobile & Personal Computing, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

ai-name-audit-deferred

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.10451

JarvisGUI paper

• Paper https://arxiv.org/abs/2609.10451

Which sources cite JarvisGUI?

• JarvisGUI

Related Agents & Tool Use, Mobile & Personal Computing, cs.AI benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• AdaPlanBench
• Adaptive Adversaries
• ADeptS-Bench
• AdversaBench