Azure OpenAI with GPT-5.6 Terra fails with "Request higher limits" in Cursor 3.15.6

[saleesh (Saleesh Sasi)]

Where does the bug appear (feature/product)?

Cursor IDE

Describe the Bug

Azure OpenAI with GPT-5.6 Terra not working in Cursor 3.15.6 — misleading “Request higher limits” error

Azure configuration

I have an Azure OpenAI resource with:

Base URL: https://.openai.azure.com
Deployment: gpt-5.6-terra

The Azure deployment is working correctly.

I tested the same Azure endpoint, API key and deployment independently using the OpenAI Python SDK:

from openai import OpenAI

client = OpenAI(
api_key=“MY_AZURE_API_KEY”,
base_url=“https://.openai.azure.com/openai/v1/”
)

response = client.responses.create(
model=“gpt-5.6-terra”,
input=“What is the difference between RSTP and MSTP?”
)

print(response.output_text)

This works successfully and returns a valid response from GPT-5.6 Terra.

Problem in Cursor

I configured the same Azure OpenAI endpoint/API key/deployment in Cursor and selected the Azure model.

Cursor returns:

Request higher limits to continue using Cursor

This is misleading because the Azure API itself is working and the same credentials/deployment successfully work outside Cursor.

Steps to Reproduce

1. Create/configure an Azure OpenAI deployment:
2. Azure OpenAI endpoint: https://.openai.azure.com
3. Deployment/model: gpt-5.6-terra
4. Verify that the Azure deployment is working independently using the OpenAI Python SDK and the
5. Responses API:
   from openai import OpenAI
   client = OpenAI( api_key=“AZURE_API_KEY”,
   base_url=“https://.openai.azure.com/openai/v1/”
   ) response = client.responses.create(
   model=“gpt-5.6-terra”,
   input=“What is the difference between RSTP and MSTP?”
   ) print(response.output_text)
6. The Python request completes successfully and returns a valid response from gpt-5.6-terra.
7. Open Cursor 3.15.6 Stable.
8. Go to Settings → Models and configure the Azure OpenAI BYOK/custom API configuration using
9. the same Azure endpoint, API key and deployment.
10. Select the configured GPT-5.6 Terra model in Cursor.
11. Send a simple prompt, for example:
    What is the difference between RSTP and MSTP?
12. Cursor fails with:
    Request higher limits to continue using Cursor

When Azure OpenAI BYOK is configured in Cursor with a valid Azure endpoint, API key, and gpt-5.6-terra deployment:

Cursor should send the request to the configured Azure OpenAI endpoint.

Screenshots / Screen Recordings

Operating System

Windows 10/11

Version Information

Environment
Cursor: 3.15.6 (Stable, User Setup)
VS Code Extension API: 1.128.0
Commit: a1f686545fd0ce8917bbd2449f733551a9bce420
OS: Windows 11 x64
Model: GPT-5.6 Terra
Provider: Azure OpenAI

For AI issues: which model did you use?

Model: gpt-5.6-terra

Provider: Microsoft Azure OpenAI

Azure deployment: gpt-5.6-terra

Does this stop you from using Cursor

Yes - Cursor is unusable

[mohitjain (Mohit)]

Hey! Your Azure setup isn’t the problem, and Cursor did receive your key. “Request higher limits to continue using Cursor” is a usage-limit block, not an Azure error.

On token-based (pooled) Team/Enterprise plans, custom-key (BYOK) requests still count toward your usage limits, so once you hit your monthly limit they get blocked with this message too. That’s also why a wrong deployment name gives the same error, the request stops before it ever reaches Azure.

Fix: use Request Limit Increase on that dialog (or ask your team admin to raise your per-user limit), or wait for the monthly reset. After that, requests will route to your gpt-5.6-terra deployment normally.

Agreed the wording is confusing when the key is valid, noted that as feedback.