CAP Benchmark

What is CAP?

CAP evaluates cross-site browser agents on 420 tasks across 108 real websites and 24 domains. Tasks require complex UI interactions and visual perception, with…

Released

2026-08-09

Evaluates

General AI, Language & Knowledge, cs.AI

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.08392

CAP paper

• Paper https://arxiv.org/abs/2608.08392
• Website https://warriorxu0302.github.io/CAP-Bench/

Which sources cite CAP?

• CAP

Related General AI, Language & Knowledge, cs.AI benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A Large-Scale Dataset and Benchmark
• A3S-Bench
• AARRI-Bench
• ABC-Bench
• ActBench