(context_length_exceeded) This model's maximum context length is 128000 tokens for gpt-4.1

[TsuyoshiUshio]

• Package Name: azure_ai_inference
• Package Version: 1.0.0b9.dist-info
• Operating System: Ubuntu (WSL2) 20.04.4 LTS
• Python Version: python 3.12.9

Describe the bug
I use 4.1 model that has long context, but received the message

To Reproduce
Steps to reproduce the behavior:

1. Create a sample app.
2. Provide a message that has more than 128000 token
3. Get the error message.

Error Message

Error during LLM analysis for emergency issue: (context_length_exceeded) This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
Code: context_length_exceeded
Message: This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
Traceback (most recent call last):
  File "/home/ushio/Code/Project/KustoAutogenExperiment/extract_emergency_issue.py", line 396, in analyze_with_llm
    response = openai_client.complete(
               ^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/ushio/Code/Project/KustoAutogenExperiment/venv/lib/python3.12/site-packages/azure/ai/inference/_patch.py", line 738, in complete
    raise HttpResponseError(response=response)
azure.core.exceptions.HttpResponseError: (context_length_exceeded) This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
Code: context_length_exceeded
Message: This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
Successfully processed ICM 626717326 for emergency issue information.
Emergency issue analysis failed: (context_length_exceeded) This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
Code: context_length_exceeded
Message: This model's maximum context length is 128000 tokens. However, you requested 1124171 tokens (124171 in the messages, 1000000 in the completion). Please reduce the length of the messages or completion.
All relevant files saved in the extracted_emergency_issues directory.

I change the max_tokens settings but result is the same.

  response = openai_client.complete(
      model=self.openai_deployment,
      messages=messages,
      max_tokens=1000000,  
      temperature=0.1 
  )

Expected behavior
In case of long context model like 4.1 we can use the context window.

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @dargilco @jhakulin @trangevi.

[github-actions]

Hi @TsuyoshiUshio, I'm an AI Support assistant here to help with your issue. While the Azure SDK team reviews your request, I wanted to provide some possible tips and documentation that might help you in the meantime.

• The error indicates that the total token count (sum of tokens in the messages plus the completion tokens) exceeds the model’s maximum context length of 128000 tokens. Even though you are using a long context model (GPT-4.1), the model enforces this upper limit.
• Setting max_tokens to 1000000 leads to requesting far more tokens than supported. The sum of the tokens in your messages and the requested completion must remain below 128000.
• Review the complete request parameters so that your messages and the completion token count together do not exceed the model’s limit. See details on model-specific parameters in the azure-ai-inference README.

Confidence Score: 0.9

The Azure SDK team will respond to your issue shortly. I hope these suggestions are helpful in the meantime. If this comment helped you, please give it a 👍. If the suggestion was not helpful or incorrect, please give it a 👎. Your feedback helps us improve!