[KongchangAI]
BenchmarkMKA

MKA

A multimodal knowledge-work benchmark jointly released by Google and Artificial Analysis, comprising 200 tasks designed by industry experts, used to evaluate the delivery quality of models in knowledge-work scenarios.

Timeline (last 90 days)

Nov 15

Google and Artificial Analysis launched MKA, which uses 200 tasks to assess agents' ability to read complex materials such as drawings and scanned documents and then complete spreadsheets and presentations; in the first batch of results, Fable 5.2 and Carbon ranked at the top.

Unverified50%
Nov 14

On MKA, a multimodal knowledge-work benchmark jointly launched by Google and Artificial Analysis, Gemini for Carbon ranked second in delivery quality, at about one-seventh the average cost of the first-place Claude Fable 5.2.

Unverified50%

All Facts (2)

Source Articles