AI API Cost Calculator
Estimate monthly AI costs for GPT, Claude and Gemini API usage, AI image generation, a customer chatbot, or paid AI seats for a team.
About this tool
AI API Cost Calculator
Estimate monthly AI costs for GPT, Claude and Gemini API usage, AI image generation, a customer chatbot, or paid AI seats for a team.
Pick the service, enter your token, image or seat volume and see what it costs per month. This page brings together 17 separate ai api cost calculator pages (AI images, Chatbot, Claude, Gemini, GPT (OpenAI), Team AI seats, Per 1,000 requests, Per user (LLM), Per conversation, Per support ticket, Per document processed, Per image generated, Per transcription minute, Per blog post, Per content piece, Subscription vs API, Batch API savings), Team AI seats) so you can compare them without hunting for the right one.
How to use AI API Cost Calculator
- 1
Open AI API Cost Calculator on AltFTool — it loads instantly in your browser.
- 2
Choose your service at the top: AI images, Chatbot, Claude, Gemini and 2 more. On a phone it is a drop-down list.
- 3
Fill in the ai api cost calculator for that service; each of the 17 options keeps its own rules, data and defaults.
- 4
Read the result, then switch service to compare. The page link keeps your service (?v=ai-image), so you can bookmark or share it.
Why use AI API Cost Calculator
All 17 in one place
AI images, Chatbot, Claude and the rest sit behind one picker, so switching service takes a tap instead of a new search.
Built for each option
Every service keeps its own rules, numbers and defaults rather than one generic template.
Shareable view
The link remembers the service you picked, so a bookmark or a shared link opens straight on it.
Common use cases
- A marketer costing 25 product-shot prompts at 4 variants each on DALL·E 3 HD before pitching the campaign budget
- Budget a RAG support bot before launch, when you know average handle length but not the token bill.
- A team comparing the monthly bill for the same workload on an Opus-tier, Sonnet-tier and Haiku-tier Claude model before committing
- Size the monthly bill for a Gemini 2.5 Flash feature before shipping it, from measured average prompt and completion lengths.
Frequently asked questions
Which service options does this tool cover?
17: AI images, Chatbot, Claude, Gemini, GPT (OpenAI), Team AI seats, Per 1,000 requests, Per user (LLM), Per conversation, Per support ticket, Per document processed, Per image generated, Per transcription minute, Per blog post, Per content piece, Subscription vs API, Batch API savings. Pick one at the top of the tool; the page link updates (for example ?v=ai-image) so you can bookmark or share that exact view.
How much does it cost to generate an image with DALL·E 3?
$0.040 per standard-quality 1024×1024 image via the OpenAI API, rising to $0.080 for standard wide formats (1024×1792 or 1792×1024) or HD square, and $0.120 for HD wide. A hundred standard square images therefore cost about $4 before retries.
How much does an AI chatbot cost per month?
It is set by tokens, not by seats. A four-turn RAG conversation with an 800-token system prompt and 2,000 tokens of retrieved context runs roughly 13,000 prompt and 900 output tokens; at $1 per million input and $5 per million output that is under two cents a conversation, so 50,000 conversations cost a few hundred dollars.
How much does the Claude API cost per million tokens?
On Anthropic's first-party API, Opus-tier models (Claude Opus 5, Opus 4.8) list at $5 per million input tokens and $25 per million output tokens, Sonnet-tier models at $3/$15, and Claude Haiku 4.5 at $1/$5. Claude Sonnet 5 is currently on an introductory $2/$10 rate through 2026-08-31, reverting to $3/$15 after (this figure must stay in sync with the claude-sonnet-5 preset in lib.js). Prices change with model releases and differ on Amazon Bedrock and Vertex AI, so confirm against Anthropic's pricing page before budgeting.
How is Gemini API pricing calculated?
Per million tokens, priced separately for input and output. Multiply your prompt tokens by the input rate per million and your generated tokens by the output rate per million, then add them. Output is normally several times more expensive than input, so completion length usually drives the bill more than prompt length.
How is OpenAI GPT API pricing calculated?
Per token, with separate rates for input (prompt) and output (completion) tokens quoted in dollars per 1 million tokens. Your cost per request is input tokens times the input rate plus output tokens times the output rate, divided by one million — so a 1,000-token prompt with a 500-token answer on a $1.25/$10 model costs about $0.00625.
How do I calculate the real cost per user of an AI subscription?
Divide the total monthly subscription by the number of people who actually used it in that month, not by the number of seats you bought. Twenty-five seats at 30 a month is 750, but if only 14 people logged in the real cost is about 53.57 per active user — and the 11 idle seats are 330 a month of pure waste.