ColinBuilds.com · Official OpenAI API pricing reference
GPT-4o
- Input / 1M
- $2.50
- Output / 1M
- $10.00
- Official context window
- 128,000 tokens
OpenAI gpt-4o official API pricing; input $2.50/1M, output $10.00/1M · as of June 29, 2026
Compare models
Same per-1M input/output rates as above. Enter your monthly token volumes to compare bills.
GPT-4o
- Input: 1M × $2.50 = $2.50
- Output: 1M × $10.00 = $10.00
- Total
- $12.50
Claude Haiku 4.5
- Input: 1M × $1.00 = $1.00
- Output: 1M × $5.00 = $5.00
- Total
- $6.00
Difference
$6.50
GPT-4o costs $6.50 more than Claude Haiku 4.5.
Popular comparisons
30 total featuring GPT-4o.
OpenAI’s GPT-4o is a dependable general-purpose multimodal API model: text and image in, text out, strong enough for everyday assistants, research workflows, content help, and production tool use.
What this model is
GPT-4o is OpenAI’s versatile flagship multimodal model. OpenAI describes it as a high-intelligence model that accepts text and image inputs and produces text outputs, including Structured Outputs. In practical terms, this is the model many teams compare against when asking what a reliable mainstream AI API costs for normal app work.
This page is a pricing reference for the GPT-4o API route, not a full benchmark profile. Parameter counts and benchmark scorecards are omitted here until exact official values are verified.
Pricing notes
OpenAI’s GPT-4o model page lists text token pricing at $2.50 per 1M input tokens, $1.25 per 1M cached input tokens, and $10.00 per 1M output tokens. The calculator on this page uses the standard uncached input and output rates: $2.50 input and $10.00 output per 1M tokens.
The pricing is useful as a mainstream comparison point because many cheaper models are judged against GPT-4o on cost, speed, reliability, and ecosystem support. Tool calls, search, audio, image generation, and other OpenAI products can have separate pricing, so this page is not a full OpenAI account-cost estimate.
Benchmarks and specs
OpenAI’s GPT-4o model page lists a 128,000 token context window and 16,384 max output tokens. The same page describes GPT-4o as supporting text input/output, image input, function calling, Structured Outputs, fine-tuning, and predicted outputs.
OpenAI does not publicly disclose GPT-4o parameter count on the checked model page. This page avoids unsourced benchmark claims until a specific public benchmark source is verified.
Best fit
GPT-4o is best for production assistants, customer support automation, research workflows, structured-output apps, and general API products where reliability and OpenAI ecosystem support matter more than chasing the absolute lowest token price.
Use GPT-4o as a mainstream default comparison, then check whether a cheaper specialist model is good enough for the exact job.