LLM API cost calculator
Pick a model and your monthly token volume to estimate what the LLMBase Inference API would cost.
Estimate your cost
What would your monthly volume cost?
Pick a model and a typical workload, or adjust monthly token volume, input/output ratio, and cache hit rate yourself.
$0.00
$0.00 blended / 1M tokens
Estimate only, based on the normal catalogue prices. Billing uses actual token usage. The API with location Germany (on request) adds a 20% surcharge on all token prices, including cached input, and is not included in this estimate. Long-context surcharges are not included either.
Tokens / month
Input / output ratio
85 / 15Cache hit rate
60%How the estimate is calculated
The calculator splits your monthly tokens into input and output using the ratio you set. Input tokens served from the prompt cache cost the model's cache-read price, the remaining input tokens its input price, and output tokens its output price, each per 1M tokens. If you know requests rather than tokens, enter requests per month and average tokens per request; the calculator fills in their product as tokens per month.
FAQ
Where do the prices come from?
They load live from the public Inference API model catalogue (`GET /v1/models?metadata=true`).
How are cached tokens calculated?
Cache hits use the model's cache-read price; cache writes are billed like normal input tokens.
Is the result binding?
No — it is an estimate using the normal catalogue prices. Billing uses actual token usage against your prepaid balance in USD. The API with location Germany (on request) adds a 20% surcharge on all token prices, including cached input, and is not included.
Is there a monthly base fee?
No. Pay-per-use with no monthly fees.
LLMBase Inference
Start with the Inference API
Get your API key, swap the base URL, go.
Pay-per-use · No monthly fees