LLM tokens and cost
LLM model price comparison.
Type in two or three price sets and one workload. See the monthly cost of each and the volume where a fixed monthly fee starts to pay off.
Monthly cost = fixed fee + requests x (input tokens x input price + output tokens x output price) / 1,000,000. For 2,000 input and 500 output tokens, example price sets of $1.00 / $4.00 and $0.20 / $0.80 per 1M with a $300 fixed fee on the second, the second is cheaper above 93,750 requests a month. At 100,000 requests it costs $380.00 against $400.00.
Your numbers
Result
Worked example
2,000 input and 500 output tokens per request, 100,000 requests a month. Option A (hosted API, example $1.00 and $4.00 per 1M, no fixed fee): (2,000 x 1 + 500 x 4) / 1,000,000 = $0.004000 per request, $400.00 a month. Option B (example $0.20 and $0.80 per 1M, $300 fixed): $0.000800 per request, 300 + 100,000 x $0.000800 = $380.00 a month. They cost the same where 300 + $0.000800 x r = $0.004000 x r, so r = 300 / $0.003200 = 93,750 requests a month. Below that, A is cheaper; above it, B is.
How it works
The page compares cost only. It does not judge quality, speed, rate limits or context size, and a cheaper price set only helps if its answers are good enough for your task. Test candidates on your own prompts before you switch.
The fixed monthly cost lets you compare a pay-per-token API against something with a standing fee, such as reserved capacity or self-hosting. The break-even volume between two options is (fixed fee of the one with the larger fee - fixed fee of the other) / (difference in cost per request). If the option with the higher fixed fee does not also have the lower per-request cost, there is no break-even. Take every price from the provider's pricing page for the exact model, and re-check it, since prices change. Per-request costs here use the same arithmetic as the cost calculator.
No outside source is needed: this page is plain arithmetic on the numbers you enter.
All prices on this page are examples you can overwrite, not any provider's current prices. Check your provider's pricing page before you rely on a number. Results are planning estimates, not a quote or financial advice. Everything runs in your browser. Nothing you type is sent or stored.
FAQ
How do I compare the cost of two LLMs?
Use the same workload for both: average input tokens, output tokens and requests per month, then multiply by each price set. Here the cost per request is $0.004000 against $0.000800.
What is the break-even volume?
It is the monthly request count where two options cost the same. With a $300 fixed fee on option B and a saving of $0.003200 per request, B overtakes A at 93,750 requests a month.
Does the cheapest model give the best value?
Not necessarily. This tool compares price only. Quality, latency, rate limits and context size matter, so test each candidate on your own prompts.
Where do I get the prices?
From each provider's own pricing page, for the exact model and region. Every price on this page is a placeholder you replace.