AI Persona Generator
Create a persona for a chatbot or roleplay character, as a prompt or a character card.
LLM Cost Calculator
Work out what an AI feature will cost per call, per day and per month.
One request
The tokens a typical call uses. The token counter can estimate these from real text.
Repeated context served from a prompt cache, billed at roughly a tenth of the input rate.
Your models
Replace these with the current per-million rates from your provider's pricing page.
Rates are not filled in for you on purpose — providers change them often, and a hard-coded table would quietly go stale. Your list is remembered in this browser.
Projected cost
| Model | Per call | Per day | Per month | Per year | Input / output split |
|---|
Add a model with its rates to see the projection.
Quick Tool Overview
The LLM Cost Calculator works out what an AI feature costs per call, per day, per month and per year, split between input and output so you can see which side is driving the bill. Add as many models as you like and they are compared in one table, cheapest first.
- What it does
- Projects the cost of an AI feature across models and volumes
- You give it
- Tokens per request, request volume and per-million rates
- You get back
- Cost per call, day, month and year, compared across models
- Best for
- Budgeting a feature, comparing models and unit economics
About the LLM Cost Calculator
Enter the tokens a typical request uses, how many requests you expect, and the calculator gives you the cost of one call, a day, a month and a year, split between input and output so you can see which side is actually driving the bill. Add as many models as you like and they are compared in one table, cheapest first.
Every rate is an editable field rather than a hard-coded price, and that is deliberate. Model prices change often and each provider publishes its own; a calculator with baked-in numbers is wrong within months and gives you no way to notice. Type the current per-million rates from your provider's pricing page and the arithmetic is exactly right. The rows you add are remembered in your browser, so your own model list is there next time.
Key features of the LLM Cost Calculator
Input and output split out
Output tokens usually cost several times more than input, and the split shows you which one to optimise.
Editable rates, not baked-in prices
Model pricing changes often, so every rate is a field — type the current numbers and the arithmetic is exactly right.
Side-by-side model comparison
Add as many models as you want and they are ranked in one table, cheapest first.
Four time horizons
Per call, per day, per month and per year, because a fraction of a cent per call turns into a real number at volume.
Your model list is remembered
Rows you add are kept in your browser, so your own comparison is there next time.
How to use the LLM Cost Calculator
-
1
Describe one request
Enter the input and output tokens a typical call uses. The token counter tool can estimate these from real text.
-
2
Set your volume
Enter how many requests you expect per day, or switch to monthly if that is how you think about it.
-
3
Enter the rates
Add each model with its input and output price per million tokens, copied from the provider pricing page.
-
4
Compare the totals
Read the per-call, daily, monthly and annual cost for each model, sorted cheapest first.
Frequently asked questions
Because they would be out of date. Providers change per-token pricing regularly and each publishes its own rates, so a hard-coded table quietly becomes wrong. Typing the current numbers takes seconds and the result is then genuinely accurate.
Generating tokens costs far more compute than reading them, and providers price accordingly — output is commonly three to five times the input rate. That is why a chatty system prompt is cheap and a long generated answer is not.
Shorten what you generate before you shorten what you send, since output is the expensive half. After that, prompt caching on repeated context, batching where latency does not matter, and using a smaller model for the easy requests are the usual wins.
Only in your own browser storage, so your model list is still there when you come back. Nothing is sent to a server.
Reviews of the LLM Cost Calculator
What people who use the LLM Cost Calculator think of it. Reviews come from visitors and are posted as written.
No ratings yet
Yours would be the first.
- 5 0 reviews rated 5 stars
- 4 0 reviews rated 4 stars
- 3 0 reviews rated 3 stars
- 2 0 reviews rated 2 stars
- 1 0 reviews rated 1 star
0 reviews
Newest first
No reviews yet
Be the first to rate the LLM Cost Calculator. It takes half a minute and no account.
Write the first reviewMore AI tools
Embedding Cost Calculator
Estimate the cost and vector storage of embedding a document corpus for RAG.
Token Counter
Estimate how many tokens your text costs before you send it.
Function Schema Generator
Write valid tool-calling JSON schemas without hand-editing braces.
ChatML Formatter
Convert a conversation into ChatML, Llama 3, Mistral, Gemma or Alpaca prompt templates.
Cosine Similarity Calculator
Compare two embedding vectors by cosine similarity, dot product and distance.
ChatGPT Export to Markdown
Turn conversations.json from a ChatGPT or Claude export into readable Markdown files.
Fine-Tuning Cost Calculator
Estimate what a fine-tune costs to train and run, and when it pays for itself.