Queries per month
5,000
Search or answer requests that call the generation model.
RAG cost planning
RAG Cost Calculator
Plan retrieval-augmented generation costs by modeling query volume, context tokens, and answer length. No documents or private text are collected.
Scenario assumptions
Start with these editable numeric assumptions, then adjust the calculator fields to match your own traffic and token usage.
Retrieved context tokens
2,500
Chunks, citations, and source excerpts included with each query.
Question and instructions
500 tokens
User question, system guidance, and output formatting instructions.
Output tokens per answer
700
Average generated answer length after retrieval.
RAG app model and usage
Choose a model and edit the numeric assumptions for this RAG app scenario.
OpenAIexample-openai-premiumExample price
Input price
$3.00 / 1M tokens
Output price
$12.00 / 1M tokens
Last updated: 2026-06-27View pricing source
This model uses placeholder example pricing. Verify the official source before using it for planning.
Usage assumptions
Estimate traffic and token usage for an average request.
Estimated results
Run the calculator to see usage volume and projected cost.
Add your usage assumptions, then calculate to preview estimated AI API spend.
How to read the output
Use the result as a planning estimate, then validate against provider pricing and real usage once the workflow is live.
Generation cost
RAG spend is often driven by the size of retrieved context sent into the model for every query.
Context sensitivity
Run low and high context scenarios to see how chunk count and context length affect monthly cost.
Embedding note
If your provider charges separately for embeddings, estimate indexing and query embedding costs alongside this generation estimate.
Models to consider
These options come from the central model data and are suggestions to evaluate, not claims that one model is best for every workload.
OpenAI
example-openai-premium
Text model to evaluate for retrieval answer generation.
Input $3.00 / 1M, output $12.00 / 1M.
Open modelGemini
example-gemini-pro
Text model to evaluate for retrieval answer generation.
Input $2.00 / 1M, output $8.00 / 1M.
Open modelMistral
example-mistral-large
Text model to evaluate for retrieval answer generation.
Input $2.50 / 1M, output $7.50 / 1M.
Open modelOpenAI
example-openai-fast
Text model to evaluate for retrieval answer generation.
Input $1.00 / 1M, output $4.00 / 1M.
Open modelRelated planning pages
Move between use-case calculators, provider pages, and pricing comparisons without entering private content.
RAG app FAQ
Short answers for estimating this scenario without sharing private prompts, documents, or customer data.
What affects RAG app cost the most?
The biggest drivers are query volume, retrieved context size, answer length, selected model price, and whether embeddings are billed separately.
Are retrieved documents included in the URL?
No. This calculator only uses numeric token assumptions and safe model/provider slugs, never raw document text.
How should I estimate retrieved context tokens?
Start with average chunk size multiplied by the number of chunks sent to the model, then add question and instruction tokens.