Calculator

Token, cost and saving calculator

Paste real text, estimate tokens, compare models and simulate how much you can save with cache, rules, local AI or preprocessing.

Note Prices are editable estimates. Before budgeting production, always verify the provider's official pricing.

Text, prompt, JSON, code or context

Expected output per request

Requests per month

Cacheable input 20%

Model	Provider	Context	Input / 1M	Output / 1M	Monthly cost

Reference prices in USD per million tokens, editable: adjust them to your case and always verify the provider's official rate. Automatic price update: —

Input tokens per month

Output tokens per month

Input price / 1M

Output price / 1M

Reduced by cleanup 25%

Moved to local AI or rules 35%

Before 0 EUR After 0 EUR Monthly saving 0 EUR

Model size In billions of parameters. Example: 3, 7, 14, 32.

Quantization

Output tokens

Estimated speed Tokens per second.

Approximate RAM/VRAM 0 GB

With operating margin 0 GB

Generation time 0 s

What hardware runs it?

Based on model size and quantization, this is what it needs and where it fits. Green = fits comfortably in memory; grey = too small.