[BUG] Sends effort parameter when using custom model
Preflight Checklist
- [x] I have searched existing issues and this hasn't been reported yet
- [x] This is a single bug report (please file separate reports for different bugs)
- [x] I am using the latest version of Claude Code
What's Wrong?
When I use Claude code with a custom client (AWS Bedrock backend), all prompts fail with the following error message:
API Error: 400 >{"errorCode":"CUSTOM_CLIENT","errorName":"LanguageModelService:LlmHttpClientError","errorInstanceId":"XXXX","parameters":{"modelCreator":"Optional[ANTHROPIC]","responseBody":" Optional[{\"message\":\"This model does not support the effort parameter.\"}]","modelBackendId":"Optional[BEDROCK]","errorType":"Optional[VALIDATION_EXCEPTION]","errorMessage":"Optional[This model does not support the effort parameter.]","errorCode":"Optional.empty","statusCode":"400"}}
What Should Happen?
Claude should send its response.
Error Messages/Logs
API Error: 400 >{"errorCode":"CUSTOM_CLIENT","errorName":"LanguageModelService:LlmHttpClientError","errorInstanceId":"XXXX","parameters":{"modelCreator":"Optional[ANTHROPIC]","responseBody":" Optional[{\"message\":\"This model does not support the effort parameter.\"}]","modelBackendId":"Optional[BEDROCK]","errorType":"Optional[VALIDATION_EXCEPTION]","errorMessage":"Optional[This model does not support the effort parameter.]","errorCode":"Optional.empty","statusCode":"400"}}
Steps to Reproduce
- Install latest Claude code.
- Point it to a custom model.
- Send a prompt.
Claude Model
Sonnet (default)
Is this a regression?
Yes, this worked in a previous version
Last Working Version
2.1.112
Claude Code Version
2.1.141 (Claude Code)
Platform
AWS Bedrock
Operating System
Ubuntu/Debian Linux
Terminal/Shell
Other
Additional Information
Previously fixed in https://github.com/anthropics/claude-code/issues/30795, but now fails. I only have access to the specified custom model so this failure entirely blocks me from upgrading Claude code.
This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗