Free tool
AI token counter
Paste text and count tokens for Claude, GPT-6 and Muse models — exact where the tokenizer is verified, labeled estimates everywhere else.
Count tokensLLM API cost analysis
Understand and budget your LLM API costs
Count tokens and price API calls from verified, versioned rate cards — free, no sign-up. Already paying an LLM API bill? Join the pilot for project cost reviews and budgets.
Prices verified Oct 15, 2026Text stays in your browserNo API key needed
Estimate
GPT-6.1 Sol
OpenAI API · Standard · gpt-6.1-sol
1,000 API calls/day × 30 days
Uncached input$120.002,000 tok × $2.00 / 1M = $0.004000 per call
Cache read—0 tokens
Cache write · short—0 tokens · Simple mode assumes none
Output · incl. reasoning$600.002,000 tok × $10.00 / 1M = $0.02000 per call
Per call$0.02400
Period total30 days · 30,000 API calls$720.00
Monthly budget31-day month$744.00
- Uncached input16.7%
- Cache read0.0%
- Cache write0.0%
- Output83.3%
- Standard rate (price threshold 272,000 prompt tokens) — 2,000 prompt tokens per call.
- Excludes tax. Totals are summed before rounding.
Prices verified 2026-10-15Price version 2026-10-15.1USD
01 — Free tools
Two tools, no sign-up
Free tool
LLM API cost calculator
Price a call from uncached input, cache reads, cache writes and output — then see the same usage on other models.
Calculate your cost02 — How we price
Every number says where it came from
- 01
Verified, versioned prices
Each price is checked against the vendor’s own page. Results carry the check date and price version, so a saved number never changes silently.
- 02
Exact or estimated, labeled
A count is exact only where we verified the model’s tokenizer. Everything else is marked as an estimate — even within the same vendor.
Estimated - 03
No number beats a wrong number
If a price or rule isn’t verified, the cell stays blank and says why. A blank never means free.
Unverified - 04
Priced the way you’re billed
Uncached input, cache reads, cache writes and output are separate lines — and long and short requests are tiered one by one, never averaged.
03 — Pilot
Already paying an LLM API bill?
We’re building project cost reviews for small teams: bring last period’s usage, see where the money went, set a budget, and check again next period.
- Import usage from a CSV template. No API keys, no proxy, no prompts — just models, token counts, dates and call counts.
- See cost by model, input, cache and output, with what’s covered and what isn’t.
- Set a budget from your real daily average and a growth assumption you choose.
- Compare up to two alternatives — another model or a different cache split — with every assumption written down.