Claude did best on a new benchmark for agents that build agents. It still passed fewer than a quarter of the tests.

AI models now power all manner of agents, from coding assistants that write and debug software to customer service syste...

By Dhallo News · September 10, 2026 at 1:44 AM IST

Claude did best on a new benchmark for agents that build agents. It still passed fewer than a quarter of the tests.

Source: The New Stack

AI models now power all manner of agents, from coding assistants that write and debug software to customer service systems The post Claude did best on a new benchmark for agents that build agents. It still passed fewer than a quarter of the tests. appeared first on The New Stack.

SOURCE: The New Stack

ORIGINAL REPORT: Click here

DHALLO EDITORIAL STATUS: Aggregated & reviewed

Have feedback on this article?Report error or suggestion

Claude did best on a new benchmark for agents that build agents. It still passed fewer than a quarter of the tests.

Read More from Technology

OpenAI gave an AI the power to block its own engineers’ code

“It could kill us all”: what Anthropic’s own researchers really think about superintelligence

watchOS 27 upgrade coming to Apple Watch this month

5 minutes. The news that matters.