[KongchangAI]
BenchmarkMKA

Multimodal Knowledge Agent

A new benchmark developed by Google in collaboration with Artificial Analysis to test model performance on multimodal knowledge work within agentic workflows.

Timeline (last 90 days)

Nov 12

Hydrogen scores 53.4% on the Multimodal Knowledge Agent benchmark and 71.6% on Agents' Last Exam v1.1.

Unverified50%
Nov 10

Google, in collaboration with Artificial Analysis, launched a new Multimodal Knowledge Agent (MKA) benchmark, on which Gemini 4 Carbon ranks second, with average cost per task more than 6 times lower than the best-performing model.

Unverified50%

All Facts (2)

Source Articles