OpenAI API Pricing Calculator

Estimate your monthly costs for OpenAI's GPT models. Compare GPT-4o, GPT-4o mini, and more.

Model Selection

Usage Parameters

Estimated Monthly Cost

Input Cost$0.00
Output Cost$0.00
Total Cost$0.00

*Estimates based on current OpenAI API pricing. Actual costs may vary.

Complete Guide to OpenAI API Costs

OpenAI's pricing model is the industry standard, but it can be complex to navigate. Whether you're building AI Chatbots or enterprise automation workflows, understanding your potential spend is the first step to ROI.

Key Takeaways

  • GPT-4o mini is the most cost-effective model for high-volume tasks.
  • GPT-4o is the flagship model for complex reasoning and multimodal inputs.
  • Batch API can save you 50% on costs for non-urgent workloads.

OpenAI API Pricing Overview

ModelInput Price ($/1M)Output Price ($/1M)Cached Input ($/1M)
Chat / Completion Models
GPT-5$1.25$10.00-
GPT-5 Mini$0.25$2.00-
GPT-5 Nano$0.05$0.40-
GPT-4o$5.00$20.00$2.50
GPT-4o mini$0.15$0.60$0.075
Audio Models
Whisper (Speech to Text)$0.006 / min$--
TTS (Text to Speech)$15.00 / 1M chars$--
TTS HD$30.00 / 1M chars$--
Fine-tuning Models
GPT-4o Fine-tuning$3.75$15.00-
GPT-4o mini Fine-tuning$0.30$1.20-
GPT-3.5 Turbo Fine-tuning$3.00$6.00-
Embedding Models
text-embedding-3-small$0.02$--
text-embedding-3-large$0.13$--
ada v2$0.10$--

GPT-4o: The Flagship Model

GPT-4o is OpenAI's most advanced model, offering multimodal capabilities (text, audio, image) and faster performance. It's ideal for complex tasks that require high intelligence.

GPT-4o mini: Cost-Effective Intelligence

GPT-4o mini is designed for speed and efficiency. It's significantly cheaper than GPT-3.5 Turbo while offering better performance, making it perfect for high-volume applications like AI Email Assistants.

FAQ

OpenAI API Pricing FAQs

Got questions? We've got answers. Here are the most common questions we get from potential clients.

OpenAI charges per 1,000 tokens (approx. 750 words). You are billed for both input tokens (what you send) and output tokens (what the model writes).