In what unit is Azure OpenAI quota assigned to deployments?
Why D is the answer
Tokens per minute
Quota is per subscription, region, model and deployment type, in tokens per minute.
In detail
- Quota is assigned in tokens per minute.
- Rate limits follow from the TPM you assign.
Use case
An illustrative scenario, not a real incident.
A team assigns 100,000 tokens per minute across its deployments of one model, and sets each by need.
See it in different ways
flowchart LR A["Subscription quota per region, model and type"] --> B["Deployment A: TPM"] A --> C["Deployment B: TPM"] B --> D["Requests over the limit get throttled"]
stateDiagram-v2 [*] --> ModelQuota ModelQuota --> DeploymentA: assign TPM ModelQuota --> DeploymentB: assign TPM DeploymentA --> ModelQuota: reduce TPM to free capacity DeploymentB --> Throttled: peak above its limit
mindmap
root(("Model quota in TPM"))
n1["Deployment A"]
n2["Most traffic"]
n3["Deployment B"]
n4["Testing"]Key point: Quota is tokens per minute.
Why the other options are wrong
- A. Users per subscription
- User counts are not the unit.
- B. Prompts per month
- Monthly prompts are not the unit.
- C. Gigabytes of storage per day
- Storage is not the unit.

