GPT-5 max tokens not supports

[lx1054331851]

App Version

3.25.11

API Provider

LiteLLM

Model Used

gpt-5

Roo Code Task Links (Optional)

LiteLLM streaming error: 400 litellm.BadRequestError: AzureException BadRequestError - Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.. Received Model Group=gpt-5
Available Model Group Fallbacks=None

🔁 Steps to Reproduce

LiteLLM streaming error: 400 litellm.BadRequestError: AzureException BadRequestError - Unsupported parameter: 'max_tokens' is not supported with this model. Use 'max_completion_tokens' instead.. Received Model Group=gpt-5
Available Model Group Fallbacks=None

💥 Outcome Summary

expected use gpt-5 model nomally

📄 Relevant Logs or Errors (Optional)

[roomote]

Hi! I'm here to help out the maintainers and am going to see if I can fix this issue. The problem appears to be that GPT-5 requires using max_completion_tokens instead of max_tokens when making API calls through LiteLLM. I'll work on updating the code to handle this model-specific parameter requirement. Thanks for reporting this!

[roomote]

I've successfully implemented a fix for this issue! 🎉

The problem was that GPT-5 models require the use of max_completion_tokens instead of the deprecated max_tokens parameter when making API calls through LiteLLM.

Solution implemented:

• Added detection for GPT-5 model variants (gpt-5, gpt5, GPT-5, gpt-5-turbo, gpt5-preview, etc.)
• Updated both createMessage and completePrompt methods in the LiteLLM provider to use max_completion_tokens for GPT-5 models
• Maintained backward compatibility by continuing to use max_tokens for non-GPT-5 models
• Added comprehensive test coverage to ensure the fix works correctly

Pull Request: #6980

All CI checks have passed successfully. The fix is ready for review and merge.