OpenAI frontier models
GPT-5 pricing
Use this when you reference the general GPT-5 endpoint: $1.25/M input, $10/M output, $0.125 cached.
Last pricing check: Mar 13, 2026
Why teams choose this model
Scenario planning
Realistic cost examples
Numbers use GPT-5 pricing
Bulk summarization queue
Teams summarize support tickets or documents using GPT-5’s balanced pricing.
Per request
$0.0059
Per month
$528.75
Tokens sent
108,000,000
Workflow automation
Automation flows rely on GPT-5 to generate emails, updates, and QA steps.
Per request
$0.0094
Per month
$281.25
Tokens sent
57,000,000
General copilot
Internal chat assistants handle everyday questions without touching the pricier tiers.
Per request
$0.0077
Per month
$461.25
Tokens sent
96,000,000
Compare with
FAQs
Why keep GPT-5 separate from GPT-5.1?
Some teams pin to the stable release tag (5.1) while others use GPT-5 as the evergreen alias. We expose both so you can mirror whichever endpoint you call.
How should I treat context?
GPT-5 inherits the <270K tier, so TokenTally models a 270K max context like other non-Pro GPT-5 variants.
When should I move up to GPT-5.2 or 5.4?
Use GPT-5 for broad workloads; step up when you need higher accuracy, safety, or long-context reasoning.
Pricing sources
- https://developers.openai.com/api/docs/pricing
Checked Mar 13, 2026