BenchmarkMultimodal Knowledge Agent Benchmark
Multimodal Knowledge Agent Benchmark
A benchmark for evaluating the multimodal knowledge application capabilities of AI agents, covering visual file processing and deliverable generation.
Timeline (last 90 days)
Nov 15
Hydrogen leads on the OS World 2.1 offline computer-use test, and leads Sonnet and GPT 6.1 Tera on Agent's Last Exam and Multimodal Knowledge Agent Benchmark, respectively.
Unverified50%