v2.1.68

2026-03-04 · Source: Claude Code Changelog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, quick

Summary

Opus 4.6, the latest model version, is now the default for Max and Team subscribers, operating at a "medium effort" setting. This setting is described as an optimal balance between processing speed and thoroughness for most tasks. Users can adjust this effort level at any time using the `/model` command. Additionally, the "ultrathink" keyword has been re-introduced, allowing users to specifically enable a "high effort" setting for individual turns. Older Opus models, specifically Opus 4 and 4.1, have been removed from the Claude Code first-party API, with any users who had these versions pinned automatically migrated to Opus 4.6.

Key takeaway

For AI Architects and MLOps Engineers managing Claude deployments, your Max and Team subscribers will now default to Opus 4.6 at medium effort. You should inform your teams about the re-introduction of the "ultrathink" keyword for high-effort tasks and the `/model` command for effort adjustments. Ensure any legacy integrations relying on Opus 4 or 4.1 are updated, as those models are now deprecated and users automatically migrated to Opus 4.6.

Key insights

Opus 4.6 now defaults to medium effort, with "ultrathink" enabling high effort for specific turns.

Principles

• Balance speed and thoroughness
• Offer granular control over model behavior

In practice

• Use `/model` to adjust effort
• Apply "ultrathink" for high-effort turns

Topics

• Opus 4.6
• Model Effort Control
• Claude API
• Ultrathink Keyword
• Model Versioning

Best for: AI Architect, MLOps Engineer, NLP Engineer, AI Engineer, Machine Learning Engineer, Prompt Engineer

Related on AIssential

• Introducing Claude Opus 4.8 — Claude Opus 4.8 significantly enhances agentic capabilities, honesty, and alignment, offering improved performance and new features at existing or reduced costs.
• [NOTICE] This feed is no longer maintained — Recent Claude Code updates enhance performance, security, and user experience, introducing Opus 4.7 and advanced code review.
• Best practices for using Claude Opus 4.7 with Claude Code — Opus 4.7 optimizes agentic coding through adaptive thinking, calibrated effort levels, and refined interaction patterns.
• v2.1.89 — Claude Code updates focus on performance, stability, and expanded capabilities for developers.
• I Tested All 5 Effort Levels of Claude Opus 4.7 — Claude Opus 4.7's new `xhigh` effort tier is the default for coding, but its cost-performance trade-offs vary significantly.
• My Honest Review of Claude Opus 4.6: Is It Worth the Hype? — Claude Opus 4.6 offers superior coding and reasoning capabilities with adaptive thinking and extended context windows.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Claude Code Changelog.