Users Complain About gpt-5.6 Spawning Too Many Subagents

TAbrodi · x · 2026-07-12

A cited post complains that during a two-hour continuous conversation with gpt-5.6, the model relentlessly spawned subagents for almost everything and was "overly aggressive" in triggering skills, forcing the user to revert to gpt-5.5.

This feedback reflects the model's agentic behavior and workflow tendencies in real-world interactions: instead of simply answering, it constantly breaks down tasks and calls subagents, which can feel verbose or out of control. Replies to the original post joked about "whether it's time to switch tools."

Related event: GPT-5.6 Subagent Mechanism Sparks Cost and Over-triggering Controversy(5 posts)→

Original post →

More from Fun

Fun channel →