Quota for Azure OpenAI Service Model Changed to 8k Without Notice

Question

Hello,

When I initially set up gpt-4o, the "Tokens per Minute Rate Limit" was 150K.

Recently, upon checking the quota, it displayed as "150K of 8K", and when viewing the model's editing screen, it showed "undefinedundefine".

Here is the quota status:

When you press the edit button, it shows 150K.

Could you please explain why this situation has occurred? Additionally, is the effective quota currently 8K?

Thank you for your assistance.

Answer 1

Hello takeolexus,

It appears that the quota limits for your Azure OpenAI Service model have been adjusted. Initially, the Tokens per Minute (TPM) limit was set to 150K, but it now reflects 8K TPM. This change may be due to updates in the service's quota policies or restrictions based on your subscription type.

I attempted to reproduce the issue in my environment, but the quota remained unchanged, as shown in the screenshot below.

The effective quota currently stands at 8K TPM, which aligns with the lower limits applicable to certain subscription types, such as Monthly credit card-based plans, Azure for Students, and Free Trials. If your subscription falls into one of these categories, this will likely explain the quota reduction.

Please refer this Other offer types.

I Hope this helps. Do let me know if you have any further queries.

---

If this answers your query, do click Accept Answer and Yes for was this answer helpful.

Thank you!