LLM and RAG Cost Calculator

Model & pricing

LLM input price$ per 1M tokens

LLM output price$ per 1M tokens, usually higher

Embedding price$ per 1M tokens, 0 if local

Prices are illustrative. Check your provider for current rates and paste them in.

Source documents

Avg pages per document

Tokens per page

Number of users

Questions per user per day

Working days per month

Context (input) tokens per questionretrieved passages + prompt

Output tokens per answer

Hosting per month ($)engine VM, 0 if on device

Managed vector DB per month ($)0 if self hosted OSS

Prompt caching, share of input cached0%

Cached reads bill about 10% of the input rate. Stable system prompt, tools and docs are what you cache.

Batching, share of requests run async0%

Batch processing takes 50% off, for work that does not need an instant reply.