[FEATURE] Allow using Opus 4.6 model while 4.7 gets sorted out

Status Fixed / completed
Maintainer reply None cached
Activity 4 comments · opened Apr 17, 2026 · closed Apr 17, 2026

Preflight Checklist

  • [x] I have searched existing requests and this feature hasn't been requested yet
  • [x] This is a single feature request (not multiple features)

Problem Statement

Opus 4.7 is currently doing a significantly worse job on real coding tasks than Opus 4.6 did, even setting the effort level to max. It has been a real productivity killer.

Proposed Solution

At least for as long as Opus 4.7 is still not up to par, please re-enable setting Opus 4.6 as the model.

Alternative Solutions

_No response_

Priority

Critical - Blocking my work

Feature Category

API and model interactions

Use Case Example

_No response_

Additional Context

_No response_

View original on GitHub ↗

4 Comments

github-actions[bot] · 4 months ago

Found 3 possible duplicate issues:

  1. https://github.com/anthropics/claude-code/issues/49689
  2. https://github.com/anthropics/claude-code/issues/49601
  3. https://github.com/anthropics/claude-code/issues/49884

This issue will be automatically closed as a duplicate in 3 days.

  • If your issue is a duplicate, please close it and 👍 the existing issue instead
  • To prevent auto-closure, add a comment or 👎 this comment

🤖 Generated with Claude Code

danra · 4 months ago

Directly writing into settings as suggested here allows using Opus 4.6 again.

Yaakov-Belch · 4 months ago

Supporting this request, and proposing a longer-term framing beyond "keep 4.6 while 4.7 gets fixed."

4.7 may not be broken — it may simply be optimizing for a different goal. Its literal instruction-following, benchmark performance, and self-verification behavior serve agentic-coding workflows where precision matters more than collaboration. That's a legitimate direction.

But it comes at the cost of the interpretive, collaborative personality that defined the Haiku → Sonnet → Opus lineage through 4.6. Many of us chose Anthropic specifically for that personality — a model that understood intent, not just instructions. These are fundamentally different interaction styles, and "re-tune your prompts" is not a substitute.

The ask: rather than treating 4.6 as a temporary fallback, consider maintaining two model lines:

  • One continuing the 4.6 tradition (interpretive, collaborative, warm) for users who value intent-inference and iterative work
  • One continuing the 4.7 direction (literal, benchmark-optimized, self-verifying) for agentic/autonomous workflows

This is how other platforms handle personality-significant model changes — they branch, not replace. Deprecating 4.6 by June without a personality-compatible successor forces a large user segment onto a model that actively works against their workflow.

I cannot use Opus 4.7 for my work. Removing 4.6 will force me to switch away.

github-actions[bot] · 4 months ago

This issue has been automatically locked since it was closed and has not had any activity for 7 days. If you're experiencing a similar issue, please file a new issue and reference this one if it's relevant.