How much does GPT-4o cost per million tokens?

Q: How much does GPT-4o cost per million tokens?

GPT-4o costs $2.50 per million input tokens and $10.00 per million output tokens. Here's how that translates to real per-call costs and how it compares to alternatives.

The short answer

As of 2026-04-26, OpenAI charges:

$2.50 per 1,000,000 input tokens
$10.00 per 1,000,000 output tokens

Output is 4× the price of input. This is typical for OpenAI and Anthropic models — generation is more expensive than reading.

What that means in real money

For a typical chat exchange (1,000-token prompt, 200-token reply):

input:  1,000 / 1,000,000 × $2.50 = $0.0025
output:   200 / 1,000,000 × $10.00 = $0.002
total per call:                      $0.0045
total per 1 million calls:        $4,500

For a longer RAG-style query (10,000 input, 500 output):

input:  10,000 / 1,000,000 × $2.50 = $0.025
output:    500 / 1,000,000 × $10.00 = $0.005
total per call:                       $0.03
total per 1 million calls:          $30,000

The longer the input, the more dominant the input cost becomes — and the more your model choice matters for total spend.

Cheaper alternatives

If you're cost-sensitive, GPT-4o is mid-range:

Model	Input ($/M)	Output ($/M)	When to consider
GPT-4o mini	$0.15	$0.60	Most workloads — 17× cheaper, smaller quality gap than you'd expect
Gemini 2.5 Flash	$0.075	$0.30	Cheapest exact-tokenizer option, 1M context
Claude Haiku 4.5	$0.80	$4.00	When you want Claude's instruction-following at low cost
DeepSeek V3	$0.27	$1.10	Cheapest frontier-tier model (subject to compliance fit)

More expensive but stronger on hard prompts

Model	Input	Output	When to consider
Claude Sonnet 4.6	$3.00	$15.00	Better instruction-following on nuanced tasks
Claude Opus 4.7	$15.00	$75.00	Frontier reasoning, when output quality justifies the bill
Gemini 2.5 Pro	$1.25	$10.00	Long-context (2M tokens), multimodal

Get a live cost estimate

Paste your actual prompt into the counter. It will show exact token counts across every model and the per-call cost based on your expected output ratio.

Try this on every model

Claude Opus 4.7 $15.00/$75.00
Claude Sonnet 4.6 $3.00/$15.00
Claude Haiku 4.5 $0.80/$4.00
GPT-4o $2.50/$10.00
GPT-4o mini $0.15/$0.60
GPT-4 Turbo $10.00/$30.00
Gemini 2.5 Pro $1.25/$10.00
Gemini 2.5 Flash $0.07/$0.30
Llama 3.1 405B $3.50/$3.50
Llama 3.1 70B $0.59/$0.79
Llama 3.1 8B $0.18/$0.18
Mistral Large $2.00/$6.00
DeepSeek V3 $0.27/$1.10
Qwen 2.5 72B $0.90/$0.90
Qwen 2.5 Coder 32B $0.80/$0.80

Try the live counter →