AI apps · Usage only · unit: thousand-token bundles
OpenAI API pricing teardown: usage-based rates
Short answer
The OpenAI API charges per token with separate input, output, cached-input and service-level rates that vary by model.
Pricing mechanics
| Plan or rate | Position in ladder |
|---|---|
| Standard | Entry |
| Batch | Paid step |
| Flex | Paid step |
| Higher-speed options where supported | Top published step |
- What the docs describe
- The API sets separate input, output, cache and service-level rates, which vary by model.
- Free or trial fence
- Not stated in the linked pricing documentation.
- Overage or upgrade rule
- Pure usage: every token is billed at the model's rate for its service level.
- Cap or spend control (none stated)
- Not stated in the linked pricing documentation.
A feature or configuration fence (what a plan includes) is not the same as an abuse cap. We do not infer private costs, margins or unpublished limits.
Takeaway for founders
Show the token mix behind any estimate. A thousand-token bundle is a convenient blended unit, but input, output and cache rates move it a lot.
Load into simulator
Illustrative remix · amounts and costs are editable.
Fictional tiers inspired by the pattern above. It does not reproduce OpenAI API's plans, invoices or margins. Fictional variable cost: $0.003 per thousand-token bundle. A thousand-token bundle is a sample blended unit; the documented per-million rates are separate.
| Tier | Model | Price | Limit | Accounts |
|---|---|---|---|---|
| Sample entry | Usage only | $0.009 per unit | hard cap 20,000 thousand-token bundles | 200 |
| Sample growth | Usage only | $0.007 per unit | hard cap 100,000 thousand-token bundles | 70 |
| Sample scale | Usage only | $0.006 per unit | hard cap 400,000 thousand-token bundles | 15 |
Related guide: Hybrid vs. usage-based billing · All teardowns