What does your prompt cost?

Paste anything. See token counts and price across every major model — computed in your browser, not on a server.

0 chars0 words0 lines

Cost by model

Model	In /1M	Out /1M	Context	This input

Batch & repo estimator

Drop many files at once for a combined token and cost total. Files are read locally and never uploaded.

choose files

File	Chars	OpenAI	Claude	Gemini
No files yet

Open-model tokenizers

Exact counts for Llama 3, Mistral and Grok 1. The real tokenizer downloads once from Hugging Face, then runs offline in your browser.

model

—select a model to load its tokenizer

Prompt compressor

Strips filler, hedge words and citations to cut tokens. Rule-based, so dense technical prose may compress very little.

strip inline citations like [12, 40] aggressive — also cut hedge words

paste a prompt above

Multimodal calculator

Image, video and audio token counts using Google's published Gemini formulas.

images

video sec

audio sec

0tokens · 0 image + 0 video + 0 audio

$0.00000

Caching savings

Model the cost of sending the same prompt repeatedly with prompt caching enabled. Uses the text from the main calculator.

model

repeat calls

paste text in the main calculator first

Browser extension

See cost while you type — not after.

API

Token counts and cost for any text, across every supported model. Free to use while in beta.

POST /api/v1/tokenize

curl -X POST https://YOUR-DOMAIN/api/v1/tokenize \
  -H 'content-type: application/json' \
  -d '{"text":"Hello world","models":["gpt-5.4"]}'

Model list and live rates are public and unauthenticated at /api/v1/models

Status

beta

The API is free and open while in beta. There are no accounts or keys yet, and no published request limits — there is basic abuse protection, but we won't quote a number we can't guarantee. Build against it, but don't depend on it for anything critical until it's out of beta.

Paid tiers with committed limits and an SLA are planned. If you want to be told when they land, or you need higher volume sooner, get in touch.

Response fields

Field	Meaning
tokens	Token count for that model
method	`exact` — counted with the real tokenizer · `estimate` — character ratio, no public tokenizer exists
input_cost_usd	Cost to send this text as input
fits_context	Whether it fits that model's context window

Case study: team cost dashboard

A concept for what per-team LLM cost tracking could look like. Built as a design exercise, not a live product.

ILLUSTRATION ONLY

Every number in this dashboard is randomly generated sample data. Nothing is connected to a backend, no accounts exist, and no real usage is being tracked. It exists to show the shape of the idea.

The problem it addresses

A calculator answers "what will this cost?" before you send a prompt. It doesn't answer "what did we spend last month, which team spent it, and why is it climbing?" — which is the question that actually reaches a finance or engineering lead once a company has several LLM features in production.

Two audiences, one surface

View	Answers
Business	Spend over time, budget vs. cap, cost by team and project, which provider dominates
IT / developer	Request volume, error rate, latency, per-key rate-limit headroom, request log

What making it real would require

Authentication so each organisation sees only its own data, a database logging every API call, an SDK that wraps a customer's LLM calls to feed that log, and a scheduled job for budget alerts. That's a different product from a calculator — which is why it's shown here as a concept rather than shipped as a feature.

Open the illustrative dashboard →

FAQ

No. OpenAI and open-model counts run entirely in your browser using the real tokenizers. Nothing is uploaded, logged or stored. The only exception is if you click “get exact count” for Claude or Gemini — that sends your text to our server, which forwards it to the provider’s counting endpoint and returns just a number.

OpenAI, Llama, Mistral and Qwen publish their tokenizers as open files, so we run the real thing offline. Anthropic and Google only expose token counting as an API call that needs a secret key — a key can’t live safely in browser code, so those show a character-ratio estimate until you request an exact count. Grok, DeepSeek and Perplexity publish no tokenizer at all, so those are always estimates.

Usually within a few percent for ordinary English prose. They drift further on code, non-Latin scripts and heavily formatted text, where tokenizers behave differently. Treat estimates as a planning figure, not a billing figure.

Rates are reviewed regularly against each provider’s public pricing page, but providers change prices without notice and rates vary by region, tier and committed-use discount. Always confirm against your own billing dashboard before making a purchasing decision.

Yes — a single POST /api/v1/tokenize endpoint returns counts and cost for any text across all supported models. The free tier covers 10,000 requests a month with no card required.

It removes conversational filler and hedge words. Text that’s already dense — academic papers, technical docs, minified code — has little for it to remove, so it correctly leaves it alone. Try the citation and aggressive toggles for academic prose.

Input tokens are what you send; output tokens are what the model generates back. Output is usually priced three to five times higher. This tool measures your input — your real bill depends on how much the model writes in reply.