Free tool / cost
Make that prompt shorter.
Paste a prompt. Get a shorter version, the tokens saved and what each call costs on the models you use.
No sign-up / runs in your browser
Trim the tokens, keep the meaning.
Output costs more than input.
To work out the monthly saving.
Paste a prompt, or try the example, to see how many tokens it uses, a shorter version, and what each call costs.
Only changes that cannot alter the meaning are applied automatically. Cutting filler words, examples or whole sections changes behaviour, so those are suggestions for you to test. Always re-run your own checks on a shortened prompt.
How it works
Three steps, no sign-up.
- 01
Paste your prompt
A system prompt or a full prompt. Nothing leaves your browser.
- 02
Review the changes
Safe fixes are applied for you: extra spaces, repeated lines, wordy phrases and indented JSON examples. Riskier ideas are listed for you to decide.
- 03
Copy the shorter prompt
Check it reads right, then see the saving per call and per month on your model.
Questions people ask
Will a shorter prompt change my results?
The automatic changes (spaces, exact repeats, wordy phrase swaps and compact JSON) keep the meaning. Anything that could change behaviour is only suggested, never applied. Test before you ship either way.
How are the savings calculated?
The tool counts tokens before and after with the real OpenAI tokenizer and scales for other model families. It is a measured difference for your text, not a guess.
What is prompt caching and why does it matter?
Providers charge much less for the identical start of a prompt they have seen recently. Putting dates, ids and user details near the top breaks that, so move changing values to the end.
Is my prompt stored?
No. It is processed in your browser only.
What else cuts cost besides shorter prompts?
Fewer model calls per task, a cheaper model for easy steps, capped output length and fewer retries. Tracify shows which of these your agent spends the most on.
More free tools
- Agent log checker
Paste a log or trace. See where it failed, what was slow and what wasted tokens.
- Token counter
Paste text. See the token count and price on GPT, Claude, Gemini and Llama.
- Model price comparison
Every model's price in one table, plus what your app costs per month on each.
- Context window checker
Does my document fit in each model's memory?
Now measure the real thing.
Send a first trace and see the true cost, speed and failures of every run.