Skip to content

LLM Workflow Cost Calculator

Price a multi-step LLM pipeline (embed, retrieve, generate, validate) per run, day and month, with cached-input rates and a provider comparison.

LLM Workflow Cost Calculator

Load preset:

Presets load a starter pipeline. You can modify, add, or remove steps freely.

Pipeline steps

1
$0.000020
gpt-6-luna: $0.1/M in, $0.5/M out
2
$0.00
3
$0.00

Cost breakdown

#StepModelInputOutputInput costOutput costTotal
1Embed Querygpt-6-luna2000$0.000020$0.00$0.000020
2Rerank Chunksgpt-4.1-nano4.0K200$0.00$0.00$0.00
3Generate Answercachedclaude-sonnet-55.0K1.0K$0.00$0.00$0.00
Total per execution$0.000020

Batch projections

Daily
$0.0020
100 executions
Monthly
$0.060
3,000 executions
Yearly
$0.730
36,500 executions

Provider comparison

What if every step used the cheapest model from one provider? Shows best-case cost per provider.

ProviderCost / executionDaily (100x)Monthlyvs current
Groq$0.0010$0.105$3.15+5150.0%
OpenAI$0.0011$0.107$3.21+5250.0%
Qwen$0.0013$0.127$3.82+6270.0%
Mistral$0.0021$0.210$6.30+10400.0%
DeepSeek$0.0027$0.273$8.19+13550.0%
Google$0.0044$0.441$13.23+21950.0%
xAI$0.0076$0.760$22.80+37900.0%
Anthropic$0.011$1.07$32.10+53400.0%
Meta$0.011$1.11$33.30+55400.0%

What this tool does

The LLM Workflow Cost Calculator models multi-step AI pipelines, such as embed, retrieve, generate, and validate stages, and reports the total cost per execution. Configure each step and provider, simulate prompt caching, and see where spend concentrates before you ship, so you can size a workflow's real cost across providers in the browser.

Updated . Provided as is. Check the output before you rely on it in production.

How to use LLM Workflow Cost Calculator

  1. 1

    Add pipeline steps

    Click Add Step to build your AI workflow. For each step, set a name (e.g., 'Embed documents'), select a step type and model, and enter the expected input and output token counts.

  2. 2

    Configure caching

    Toggle the caching checkbox on steps where prompt caching applies to see the cost reduction. The tool uses provider-specific cached input pricing.

  3. 3

    Set batch volume

    Enter how many times this workflow runs per day to see daily, monthly, and yearly cost projections across all steps.

  4. 4

    Compare providers

    Review the provider comparison table to see how total workflow cost changes if you switch all steps to a different provider, helping you optimize your model choices.

Questions and answers

What is LLM Workflow Cost Calculator?
An LLM workflow passes one request through several model calls, such as embedding, retrieval, generation and validation, so its cost is the sum of the steps. This calculator prices each step with its own model and token counts, applies cached-input rates where marked, and projects daily, monthly and yearly spend.
How is this different from a token counter?
A token counter measures one text. This tool prices a whole pipeline: each step has its own model, input and output tokens and a cached flag, the total is projected over your daily executions, and each provider is compared using its cheapest model per step.
Does it send my data to a server?
No. All calculations happen in your browser using static pricing data. Nothing is sent to any server.
How current is the pricing data?
Prices come from each vendor's pricing page and are verified on the date shown on the page. The calculator covers current models from OpenAI, Anthropic, Google, xAI, DeepSeek, Mistral and Groq.
For AI agents: how to call this tool

Machine-readable contract, endpoints and examples. Humans can ignore this section.

Best Path For Builders

Browser workflow

Runs instantly in the browser with private local processing and copy/export-ready output.

Browser Workflow

This tool is optimized for instant in-browser execution with local data handling. Run it here and copy/export the output directly.

/llm-workflow-cost-calculator/

For automation planning, fetch the canonical contract at /api/tool/llm-workflow-cost-calculator.json.