VISTA (VIsual Spec-To-App Benchmark)

What is VISTA (VIsual Spec-To-App Benchmark)?

A benchmark for LLM coding agents that build web apps from visual specs. It covers five input conditions and scores structure, behavior, and visual fidelity.

Released

2026-05-22

Evaluates

Code & Software, Software & AI Compute, cs.SE

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2605.26144

VISTA (VIsual Spec-To-App Benchmark) paper and code

• Paper https://arxiv.org/abs/2605.26144
• Code https://github.com/kaboider/VISTA_Bench
• Website https://kaboider.github.io/VIS_APP/

Which sources cite VISTA (VIsual Spec-To-App Benchmark)?

• VISTA (VIsual Spec-To-App Benchmark)

Related Code & Software, Software & AI Compute, cs.SE benchmarks

• Active-SWE
• AFT-Bench
• Agent App Benchmark
• Agent Retrieval Bench
• Agent Systems Benchmark (ASB)
• Agentic Evaluation in 2D Mazes
• Agentic Security Benchmark
• AgentInfraBench