Cost per 1M Tokens Calculator
Calculate the normalized cost per one million tokens across LLM providers.
Inputs
Cheapest Option Cost
$0.25
Cheapest Model
Claude 3 Haiku
Most Expensive Cost
$15.00
Most Expensive Model
Claude 3 Opus
Step by step
Cheapest option
Claude 3 Haiku
= $0.25
Most expensive option
Claude 3 Opus
= $15.00
Cost per Model
| Model | Price / 1M Tokens | Total Cost |
|---|---|---|
| — | $0.25 | $0.25 |
| — | $0.50 | $0.50 |
| — | $0.90 | $0.90 |
| — | $3.00 | $3.00 |
| — | $3.50 | $3.50 |
| — | $10.00 | $10.00 |
| — | $15.00 | $15.00 |
How it works
Per-token pricing varies by more than 100x between the cheapest and most expensive frontier models, making model choice one of the biggest levers on AI product cost. This calculator applies your specified token volume across a reference table of major providers' current input or output pricing so you can directly compare total spend across models before committing to one.
Formula
cost = (tokens / 1000000) × price_per_million_tokens
- tokens
- Total token volume
- price_per_million_tokens
- Model provider price per 1M tokens
Frequently Asked Questions
Should I compare input or output token cost?
Compare both if your workload has a mix — output tokens are typically 2-5x more expensive than input tokens across most providers, so completion-heavy workloads are more sensitive to model choice.
Are these prices kept up to date?
These are illustrative reference prices; providers change pricing periodically, so verify current rates on official pricing pages before finalizing a budget.
Does a cheaper model always mean lower total cost?
Not necessarily — cheaper models may require more tokens (via longer prompts, retries, or chaining) to achieve the same task quality, so per-token price isn't the only variable in total cost.