OpenAI Cost Calculator

OpenAI charges per token for both input and output. Prices vary by model. As of 2024–2025: GPT-4o costs $2.50/1M input and $10/1M output. GPT-4o-mini is $0.15/$0.60. GPT-4-turbo is $10/$30. GPT-3.5-turbo is $0.50/$1.50. o1 and o1-mini are higher for advanced reasoning. This calculator helps you estimate costs before committing.

Input tokens include your prompt, system message, and any context. Output tokens are the model's response. Most applications use more input than output, but output is often priced higher per token.

GPT-4o-mini: High volume, simple tasks, cost-sensitive. GPT-4o: Balanced quality and cost, general use. GPT-4-turbo: When you need the full 128K context. GPT-3.5-turbo: Legacy, very cheap, good for drafts. o1/o1-mini: Complex reasoning, math, coding — premium pricing.

Use this calculator to compare. A 1K input + 500 output request might cost $0.01 on GPT-4o-mini but $0.02 on GPT-4o. At 10,000 requests/day, that's $300 vs $600 per month.

The batch calculator multiplies cost per request by requests per day and days. Example: 1,000 requests/day × 30 days, each with 2K input + 1K output on GPT-4o = 60M input + 30M output tokens per month. At $2.50/1M and $10/1M, that's $150 + $300 = $450/month.

Adjust the model and token counts to see how switching to GPT-4o-mini or reducing output length affects your bill.