Free Tool · Always Up to Date

OpenAI API Pricing Calculator

Estimate your monthly OpenAI API costs across GPT-4o, GPT-4 Turbo, GPT-3.5, Whisper, DALL·E, and Embeddings. Enter your usage numbers and get an instant cost breakdown.

Real OpenAI pricing (updated April 2025)
All major models covered
Daily · Weekly · Monthly estimates
Configure Your Usage Enter tokens / calls per day
GPT-4o Flagship
Input: $2.50 / 1M tokens Output: $10.00 / 1M tokens
GPT-4o mini Fast
Input: $0.15 / 1M tokens Output: $0.60 / 1M tokens
GPT-4 Turbo Flagship
Input: $10.00 / 1M tokens Output: $30.00 / 1M tokens
GPT-4 (8K) Legacy
Input: $30.00 / 1M tokens Output: $60.00 / 1M tokens
GPT-3.5 Turbo Economy
Input: $0.50 / 1M tokens Output: $1.50 / 1M tokens
o1 Reasoning
Input: $15.00 / 1M tokens Output: $60.00 / 1M tokens
o1-mini Reasoning Fast
Input: $3.00 / 1M tokens Output: $12.00 / 1M tokens
text-embedding-3-large Embeddings
Price: $0.13 / 1M tokens
text-embedding-3-small Embeddings
Price: $0.02 / 1M tokens
text-embedding-ada-002 Legacy
Price: $0.10 / 1M tokens
DALL·E 3 Standard (1024×1024) Image
Price: $0.040 per image
DALL·E 3 HD (1024×1024) Image
Price: $0.080 per image
DALL·E 2 (1024×1024) Legacy
Price: $0.020 per image
Whisper (Speech-to-Text) Audio
Price: $0.006 / minute
TTS (Text-to-Speech) Audio
Price: $15.00 / 1M characters
TTS HD (Text-to-Speech) Audio HD
Price: $30.00 / 1M characters
Token Counter Helper Paste text to estimate tokens
0 tokens
0 words
0 characters
0 pages
Cost Summary Live
Estimated Total
$0.00
per day
Enter usage numbers above to see your cost breakdown.
$0.00 Cost per Day
$0.00 Cost per Month
0 Models Active
$0.00 Cost per Year
Model Comparison

OpenAI Model Pricing — Full Reference

All current OpenAI API prices in one place. Prices are per 1M tokens unless noted.

Model Category Input Price Output Price Context Best For
GPT-4o Flagship $2.50 / 1M $10.00 / 1M 128K General purpose, vision, best value at scale
GPT-4o mini Fast $0.15 / 1M $0.60 / 1M 128K High-volume, low-latency tasks
GPT-4 Turbo Flagship $10.00 / 1M $30.00 / 1M 128K Complex tasks, vision
GPT-4 (8K) Legacy $30.00 / 1M $60.00 / 1M 8K Legacy workloads
GPT-3.5 Turbo Economy $0.50 / 1M $1.50 / 1M 16K Budget chat, fine-tuning base
o1 Reasoning $15.00 / 1M $60.00 / 1M 200K Math, science, complex reasoning
o1-mini Reasoning Fast $3.00 / 1M $12.00 / 1M 128K Faster reasoning at lower cost
text-embedding-3-large Embeddings $0.13 / 1M — 8K High-accuracy search & RAG
text-embedding-3-small Embeddings $0.02 / 1M — 8K Budget embeddings
text-embedding-ada-002 Legacy $0.10 / 1M — 8K Legacy embedding workloads
DALL·E 3 Standard Image $0.040 / image — — Standard quality image generation
DALL·E 3 HD Image $0.080 / image — — HD quality image generation
DALL·E 2 Legacy $0.020 / image — — Budget image generation
Whisper Audio $0.006 / min — — Speech-to-text transcription
TTS Audio $15.00 / 1M chars — — Text-to-speech standard
TTS HD Audio HD $30.00 / 1M chars — — High-definition speech synthesis
FAQ

OpenAI Pricing — Common Questions

OpenAI charges per token. A token is roughly 4 characters or about 0.75 words in English. Both your input (the prompt) and the model's output (the completion) are billed separately. For example, with GPT-4o you pay $2.50 per million input tokens and $10.00 per million output tokens.
Approximately 1,333 tokens. The rule of thumb: 1 token ≈ 0.75 words, so 1,000 words ≈ 1,333 tokens. Code tends to tokenize differently — it often uses more tokens per line than prose due to whitespace and syntax characters.
GPT-4o mini is the most affordable high-capability chat model at $0.15/1M input tokens and $0.60/1M output tokens. For budget use cases with lower quality requirements, GPT-3.5 Turbo ($0.50/$1.50 per 1M tokens) is also an option. For embeddings, text-embedding-3-small at $0.02/1M tokens is extremely cost-effective.
Yes. Chat models (GPT-4o, GPT-4 Turbo, etc.) bill separately for prompt tokens (input) and completion tokens (output). Output tokens are typically 3–4× more expensive than input tokens. Embedding models, Whisper, DALL·E, and TTS only have a single price since they don't produce token-based output in the same way.
Indirectly yes. The context window is the maximum number of tokens a model can process in one request (input + output combined). If you send large system prompts or long conversation histories, all of those tokens are billed as input. Keeping prompts concise is the most effective way to reduce costs on high-volume workloads.
It depends on traffic and message length. A rough example: if each conversation averages 500 input tokens + 200 output tokens, and you handle 10,000 conversations/day — that's 5M input tokens + 2M output tokens/day. At GPT-4o rates: (5 × $2.50) + (2 × $10.00) = $32.50/day → ~$975/month. Use our calculator above with your actual numbers for a precise estimate.
OpenAI offers a Batch API that provides 50% discounts on supported models in exchange for up to 24-hour turnaround times — ideal for offline processing, data enrichment, or any non-real-time workload. For very high-volume enterprise use, OpenAI also offers custom agreements. The prices in this calculator reflect the standard pay-as-you-go API rates.
GPT-4o is a general-purpose multimodal model optimized for speed and cost across text, image, and code tasks. o1 is a dedicated reasoning model that uses chain-of-thought internally to solve harder math, science, and coding problems. o1 is significantly more expensive ($15/$60 per 1M tokens) and slower, but outperforms GPT-4o on complex reasoning benchmarks. Use GPT-4o for most tasks; use o1 only when reasoning depth is essential.

Ready to Automate Your AI Workflows?

ShipWorkflow builds the systems that connect your OpenAI usage to real business outcomes — pipelines, automations, and tools that scale.

Talk to Us
✓ Summary copied to clipboard!
Chat on WhatsApp