Feedback: Adaptive thinking in Opus 4.6 is excellent

Resolved 💬 2 comments Opened Feb 7, 2026 by marcoantoniofassa Closed Mar 8, 2026

Feedback

The adaptive thinking feature in Opus 4.6 — where the model decides on its own how much thinking effort to apply — is a huge UX improvement.

It feels noticeably better in real-world usage: fast when the task is simple, deeper when it needs to be. No manual effort tuning needed. The responses feel more natural and appropriately calibrated.

Great work on this, team! 👏

---

_Submitted by a daily Claude Code power user (CTO, using it for automation, MCP integrations, n8n, Supabase, WhatsApp bots, and more)._

View original on GitHub ↗

This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗