Explanation
Why the waste happens and who it affects.
These workloads often run intermittently, with lower throughput and shorter usage windows. When PTUs are assigned to such environments, the fixed hourly billing generates continuous cost with little utilization. Switching non-production workloads to PAYG aligns cost with actual usage and eliminates the overhead of managing PTU quota in low-stakes environments.
Billing model
The pricing dimensions that drive this cost.
PAYG charges per input and output token, making it cost-efficient for sporadic or low-volume workloads. PTUs incur fixed hourly charges regardless of utilization. Using PTUs in non-production environments typically results in paying for unused capacity.
How to detect
4 checks to find it in your estate.
- Identify OpenAI deployments in dev, test, QA, or sandbox environments running on PTUs
- Review utilization patterns showing intermittent or low-throughput activity
- Evaluate whether non-production environments require dedicated throughput or low-latency guarantees
- Check for PTU deployments where performance requirements could be met by PAYG consumption
How to fix
4 ways to remove the waste.
- Migrate non-production OpenAI deployments from PTUs to PAYG to align cost with usage
- Create environment-based deployment standards specifying PAYG as the default for non-prod workloads
- Implement workload reviews to validate whether PTU performance guarantees are truly required
- Periodically assess non-production usage patterns to prevent unnecessary PTU provisioning
Documentation
Vendor references for pricing and configuration.