NEWS

Görsel, IBM'nin Cuga, Hermes ve Deep Agents'ı hakkında, GLM-5 modelinin %30 ile %81 arasında bir tamamlanma oranına sahip olduğunu belirten bir haber başlığını göstermektedir.

IBM: Agent scaffolding on the same model boosted task completion from 30% to 81%

IBM tested the CUGA, Hermes, and Deep Agents scaffolds in the AppWorld benchmark using the GLM-5.3-flash model. The results showed that selecting the correct scaffold significantly increased task success.

This is a news summary compiled for the AiPulse newsletter — not the original article. The original article is at aihot.news.

Read the original article AI Critique → English (automatic translation)

Back to the issue