Issue Station
Sign In
Tools hereToolbox

claude-code - 💡(How to fix) Fix [BUG] Model mixes up East Asian languages [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#51665•Fetched 2026-04-22 07:56:10
View on GitHub
Comments
1
Participants
2
Timeline
4
Reactions
0
Author
yolandaycaoyolandaycao
Participants
github-actions[bot]github-actions[bot]yolandaycaoyolandaycao
Timeline (top)
labeled ×3commented ×1

Error Message

Error Messages/Logs

All responses in Simplified Chinese, every turn, without exception.

RAW_BUFFERClick to expand / collapse

Preflight Checklist

  • I have searched existing issues and this hasn't been reported yet
  • This is a single bug report (please file separate reports for different bugs)
  • I am using the latest version of Claude Code

What's Wrong?

Language setting and system prompt both set to Chinese, but model repeatedly outputs Korean

Environment:

  • Claude Code CLI
  • Model: claude-sonnet-4-6[1m]
  • language: "Simplified Chinese" set in ~/.claude/settings.json
  • System prompt explicitly states: "Always respond in Simplified Chinese"

Bug: Despite both the language setting and the system prompt requiring Simplified Chinese, the model repeatedly produces responses in Korean instead. This happens unpredictably — sometimes within the same conversation it switches between Chinese, Korean, and English with no trigger from the user.

Impact: The language setting is effectively non-functional for Chinese users. Every session requires manual policing of the output language. This is a critical UX failure for non-English locales.

What Should Happen?

It should keep outputing in Chinese, but it keeps outputting Korean. It realizes it but cannot act correctly. Happens every day every session

Error Messages/Logs

Steps to Reproduce

Steps to reproduce:

  1. Set "language": "Simplified Chinese" in ~/.claude/settings.json
  2. Add system instruction: "Always respond in Simplified Chinese"
  3. Have a multi-turn conversation
  4. Observe model randomly outputting Korean

Expected behavior: All responses in Simplified Chinese, every turn, without exception.

Actual behavior: Model intermittently switches to Korean. User must repeatedly correct it. Even after correction, it reverts to Korean in subsequent turns.

Claude Model

Opus

Is this a regression?

No, this never worked

Last Working Version

No response

Claude Code Version

2.1.116

Platform

Anthropic API

Operating System

macOS

Terminal/Shell

Terminal.app (macOS)

Additional Information

<img width="1133" height="706" alt="Image" src="https://github.com/user-attachments/assets/5692d2ad-15c9-4c95-b399-dc0aa2571de7" />

extent analysis

TL;DR

The issue can be mitigated by investigating the language setting configuration in the Claude Code CLI and the system prompt to ensure they are correctly set to Simplified Chinese.

Guidance

  • Verify that the ~/.claude/settings.json file is correctly formatted and that the "language" setting is indeed set to "Simplified Chinese".
  • Check the system prompt to ensure it is correctly set to "Always respond in Simplified Chinese" and that there are no conflicting instructions.
  • Test the model with a single-turn conversation to see if the issue persists, and then try a multi-turn conversation to observe if the language switching occurs.
  • Consider testing with a different model or version to isolate if the issue is specific to the claude-sonnet-4-6 model or a more general problem.

Example

No code snippet is provided as the issue seems to be related to configuration and model behavior rather than code.

Notes

The issue seems to be related to the model's language output, which may be influenced by various factors such as the model's training data, configuration, or the system prompt. Further investigation is needed to determine the root cause.

Recommendation

Apply workaround: Since the issue is not a regression and has never worked, it's recommended to try different configurations and test with various models to find a workaround that ensures the model outputs in Simplified Chinese consistently.

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…
#api#embedding generation#cache error#pipeline error#runtime 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

claude-code - 💡(How to fix) Fix [BUG] Model mixes up East Asian languages [1 comments, 2 participants]