Azure OpenAI model

GPT-4o cost on Azure OpenAI

Flagship multimodal model for general-purpose chat, vision, and retrieval.

Reference Cloud Calculator GPT-4o
Open the calculator All models
Illustrative pricing. Rates for GPT-4o are representative values for East US public Azure, last verified 1 Jun 2026. Verify against the Azure OpenAI pricing page before budgeting.

GPT-4o at a glance

Large context window with text and image input. The default choice when you need broad capability without paying the top reasoning tier.

Input tokens $0.0025 per 1K
Output tokens $0.01 per 1K
PTU floor / mo $2,448 15 PTU minimum
Example request $0.013 2K in / 800 out

Best for

  • Customer-facing chatbots
  • Multimodal (text + image) workloads
  • RAG over mixed content

Estimate your bill

Run your own token counts and volume.

The calculator opens pre-set to GPT-4o. Enter your input/output token sizes and monthly requests to see pay-as-you-go cost and the PTU break-even for your workload.

Open the calculator with GPT-4o

Frequently asked questions

GPT-4o cost on Azure OpenAI.

How much does GPT-4o cost on Azure OpenAI? GPT-4o is priced at $0.0025 per 1,000 input tokens and $0.01 per 1,000 output tokens (illustrative East US rates, verified 1 Jun 2026). A request with 2,000 input and 800 output tokens costs about $0.013.
When is provisioned throughput (PTU) cheaper than pay-as-you-go for GPT-4o? Once steady monthly usage passes roughly 188,308 requests per month at typical token sizes, the PTU floor of $2,448 per month beats pay-as-you-go token spend. Below that, pay-as-you-go is cheaper — and PTU should only be committed if your tokens-per-minute throughput justifies the minimum deployment.
What is the minimum PTU deployment for GPT-4o? 15 provisioned throughput units, about $2,448 per month on a one-month reservation (illustrative). Verify current minimums and rates against the Azure pricing page before committing.