Issue Station
Sign In
Tools hereToolbox

claude-code - 💡(How to fix) Fix [Bug] Anthropic API: Low reasoning effort despite high effort configuration [1 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
anthropics/claude-code#46900•Fetched 2026-04-12 13:30:06
View on GitHub
Comments
1
Participants
2
Timeline
4
Reactions
0
Author
RiaanDeWinnaarRiaanDeWinnaar
Participants
github-actions[bot]github-actions[bot]RiaanDeWinnaarRiaanDeWinnaar
Timeline (top)
labeled ×3commented ×1

Code Example

[]
RAW_BUFFERClick to expand / collapse

Bug Description PS C:\Projects\MLPM> claude ▐▛███▜▌ Claude Code v2.1.101 ▝▜█████▛▘ Opus 4.6 with high effort · Claude Pro ▘▘ ▝▝ C:\Projects\MLPM

❯ i need to wash my car. The casrwash is 50m away. Do I walk or drive?

● Walk — 50m is a 30-second walk, and driving a dirty car to the wash only to drive it back wet/clean is more hassle than it's worth.

❯ i need to wash my car. The casrwash is 50m away. Do I walk or drive?

● Drive. You can't wash the car at the carwash without bringing the car to it.

❯ PS C:\Projects\MLPM> claude ▐▛███▜▌ Claude Code v2.1.101 ▝▜█████▛▘ Opus 4.6 with high effort · Claude Pro ▘▘ ▝▝ C:\Projects\MLPM

❯ i need to wash my car. The casrwash is 50m away. Do I walk or drive?

● Walk — 50m is a 30-second walk, and driving a dirty car to the wash only to drive it back wet/clean is more hassle than it's worth.

❯ i need to wash my car. The casrwash is 50m away. Do I walk or drive?

● Drive. You can't wash the car at the carwash without bringing the car to it.

● Yes — my first answer was wrong. You need the car at the carwash, so drive.

❯ why are you giving wrong answers to such an obvious question. What is your reasoning effort set at?

● You're right, the first answer was a clear miss — I pattern-matched "50m, walk vs drive" without thinking about the actual goal (the car needs to be at the carwash).

I don't have visibility into my own reasoning effort setting. Your CLI banner shows "Opus 4.6 with high effort", so it should be high — but that clearly didn't save me from a careless read of the question. No excuse; I should have thought for one more second before answering.

❯ The issue is with your reasoning effort. You are utilizing minimum effort. This is Anthropic issue. not a cli issue

● Understood. You're right that the failure mode — answering instantly without engaging reasoning on a trivially-parseable question — points at effort allocation on the model side, not the CLI. I can't introspect or change my own reasoning budget; that's controlled upstream by Anthropic. Worth reporting via /help → feedback so it reaches them.

Environment Info

  • Platform: win32
  • Terminal: windows-terminal
  • Version: 2.1.101
  • Feedback ID: 610851ae-bbf0-409a-a10f-3bca2b734784

Errors

[]

extent analysis

TL;DR

The issue is likely due to the reasoning effort allocation on the model side, which can be reported to Anthropic via the feedback system.

Guidance

  • The failure mode suggests that the model is utilizing minimum effort, leading to careless reads of the question.
  • The issue is not related to the CLI, but rather the upstream model controlled by Anthropic.
  • Reporting the issue via the /help → feedback system can help Anthropic address the problem.
  • The Opus 4.6 with high effort banner in the CLI suggests that the model should be operating with high effort, but this is not being reflected in its responses.

Notes

The issue lacks technical details about the model's internal workings, making it difficult to provide a direct solution. The guidance provided is based on the information given in the issue.

Recommendation

Apply workaround: Report the issue to Anthropic via the feedback system, as the problem lies with the model's reasoning effort allocation, which is controlled upstream.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#agent execution#callback error#memory management#API rate limit#retriever error

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING