Estimates assume medium reasoning effort, no batch discounts and no prompt caching. Turn on
prompt caching and the input side typically drops 50–90% —
run your own mix in the free calculator preloaded with GPT-4o.
Prices were last verified on August 20, 2026 against OpenAI's official pricing page. Providers reprice
frequently — see our methodology for exactly how estimates are computed,
and always confirm final rates with the provider before budgeting.
GPT-4o FAQ
How much does GPT-4o cost?
GPT-4o charges $2.50 per million input tokens and $10 per million output tokens, with cached input reads at $1.25/M. Batch mode cuts both rates by 50% for asynchronous workloads.
What is the context window of GPT-4o?
GPT-4o supports a context window of 128,000 tokens (128K) — roughly 96,000 English words or about 192 pages of text in a single request.
Does GPT-4o support images, audio or reasoning?
GPT-4o offers vision & image understanding, audio input. Flagship multimodal model. Cached input is 50% of base input price.
How do I estimate my GPT-4o bill?
A typical workload of 10K input and 2K output tokens costs about $0.0450 per call with GPT-4o. At 1,000 requests per day that is roughly $1369.69 per month. Use the free AITokenCalculator to run your exact numbers.
