Tokens in and out
Text models bill input and output tokens at different rates. Output is usually several times more expensive.
Current HiAPI rates
AI Token & Model Cost Calculator
Estimate per-request and monthly API spend for text, image, and video models, then copy an example request that matches the selected pricing tier.
Prices come from HiAPI's current public billing configuration.
Build your monthly estimate
Pick a model type, then enter your usage. No sign-in required.
Model21 models
Example request
A format example for the selected model and tier: the parameters match the pricing tier, but the concrete token, prompt, and image values are placeholders, so the cost is not reproduced. Set HIAPI_API_KEY to a key from your dashboard and replace any example.com URLs with your own files.
curl https://api.hiapi.ai/v1/chat/completions \
-H "Authorization: Bearer $HIAPI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-opus-4-8",
"messages": [
{
"role": "user",
"content": "Summarize the key points of this document in three bullet points."
}
],
"max_tokens": 500
}'Go from estimate to first call
- 1Create an API key
Free to create. Needed to run the request above.
Create API key - 2Make your first call
Run the request or test the model in the playground.
- 3Add credits
Pay as you go. 2,000 credits = $1, with no monthly fee.
Covers this month's estimate: about 206,471 credits.
Add credits
How to use the calculator
01Pick a model
Choose text, image, or video, then search for the model you plan to call.
02Enter your usage
Type token counts or output settings and your monthly volume. Results update as you type.
03Copy the request
Take the matching cURL or Python request and run it with your API key.
What drives AI API cost?
Prompt caching
Repeated prompt prefixes are billed at a discounted cache-read rate. Some models also charge for writing to the cache.
Resolution and duration
Image and video models price by output configuration; video often bills per second of output.
AI API cost calculator FAQ
How accurate is this estimate?
It uses current public HiAPI rates. Actual charges depend on real token counts, account group, and the request log, which stays authoritative.
How do I count tokens for my prompt?
As a rough guide, 1,000 English tokens is about 750 words. The usage field in each API response reports the exact count.
Does the calculator include prompt caching?
Fields appear only when the current context tier publishes the corresponding cache rate. Writes use the default or 5-minute rate; 1-hour caching is not estimated. Read and write shares are assumptions and the example request does not guarantee a cache hit.
Why does the price change with long prompts?
Some models charge higher rates once input passes a context threshold. The calculator switches to that tier automatically and tells you when it does.
How do credits relate to dollars?
2,000 HiAPI credits equal $1. The calculator shows the monthly estimate in both so you know how much to top up.
Do I need an account to use it?
No. The calculator is free and needs no sign-in. You only need an API key to send the generated request.