AI API Cost Calculator.
Claude, GPT and Gemini models compared side by side for your actual request volume and token usage.
Prices last verified: 2026-07-18
This calculator compares per-million-token pricing across Claude, GPT and Gemini models side by side for your own request volume — 3 model families, separated into input and output token rates, since output tokens are typically priced several times higher than input.
Enter your monthly volume to see the real cost difference between models.
Why output tokens usually dominate the bill
Every major provider charges input and output tokens at different rates, and output is almost always priced several multiples higher than input — which means a chatty, verbose model response costs disproportionately more than a longer prompt sent to a concise one.
This is the single most common miscalculation teams make estimating AI costs: they budget based on prompt size and ignore that response length is usually the bigger lever. The table above lets you see exactly where your specific input/output mix lands across all three model families at once.
Keep exploring
Frequently asked questions
How is AI API cost actually calculated?+
Providers charge separately for input tokens (what you send: prompt, context, instructions) and output tokens (what the model generates), each at its own per-million-token rate — output tokens are typically priced several times higher than input tokens, so response length usually matters more to cost than prompt length.
Why do prices vary so much between models from the same provider?+
Smaller, faster models (e.g. a "Flash" or "Haiku"-tier model) cost a fraction of a flagship model's price because they use less compute per token
— the practical strategy most teams use is routing simple, high-volume tasks to a cheap model and reserving the expensive flagship for tasks that genuinely need it.
Is this pricing guaranteed to be current?+
No — AI API pricing changes frequently as providers release new models and adjust rates. This tool shows a "prices last verified" date for exactly this reason; always confirm current pricing directly with the provider before budgeting a production workload.
Does this account for caching or batch API discounts?+
No — this calculates standard per-token pricing only. Prompt/context caching and batch processing can cut costs significantly (sometimes 50–90%) for the right workload pattern, so treat this tool's output as a ceiling estimate, not the lowest achievable cost.
Which model is cheapest for my use case?+
It depends entirely on your actual token volume and mix — this calculator highlights the cheapest option for whatever request volume and token counts you enter, since the ranking can change between a high-input-low-output workload and the reverse.
How is AI API cost actually calculated?
Providers charge separately for input tokens (what you send: prompt, context, instructions) and output tokens (what the model generates), each at its own per-million-token rate — output tokens are typically priced several times higher than input tokens, so response length usually matters more to cost than prompt length.
More questions (4)
Why do prices vary so much between models from the same provider?
Smaller, faster models (e.g. a "Flash" or "Haiku"-tier model) cost a fraction of a flagship model's price because they use less compute per token
— the practical strategy most teams use is routing simple, high-volume tasks to a cheap model and reserving the expensive flagship for tasks that genuinely need it.
Is this pricing guaranteed to be current?
No — AI API pricing changes frequently as providers release new models and adjust rates. This tool shows a "prices last verified" date for exactly this reason; always confirm current pricing directly with the provider before budgeting a production workload.
Does this account for caching or batch API discounts?
No — this calculates standard per-token pricing only. Prompt/context caching and batch processing can cut costs significantly (sometimes 50–90%) for the right workload pattern, so treat this tool's output as a ceiling estimate, not the lowest achievable cost.
Which model is cheapest for my use case?
It depends entirely on your actual token volume and mix — this calculator highlights the cheapest option for whatever request volume and token counts you enter, since the ranking can change between a high-input-low-output workload and the reverse.
Cutting the bill
The practical cost-optimisation strategy most production systems use isn't picking one model — it's routing by task: cheap, fast models (the "Flash," "Haiku" or "Luna"-tier options here) handle high-volume, simple tasks, while an expensive flagship model is reserved for the smaller share of requests that genuinely need its extra capability.
This calculator's per-model comparison is exactly the input to that kind of routing decision.
None of this includes prompt/context caching or batch processing, both of which can cut costs significantly for the right workload pattern — treat the numbers above as a ceiling to budget against, not the lowest achievable cost.
See AI automation for how we typically architect model selection and cost control into a production system, and custom software for the integration layer around it.
- Cite this
- shakalakaa (Plixitt Solutions). “AI API Cost Calculator — Claude, GPT & Gemini.”
- https://shakalakaa.my/tools/ai-api-cost-calculator · Updated 2026-10-15
- Licensed under CC BY 4.0.
shakalakaa (Plixitt Solutions). "AI API Cost Calculator — Claude, GPT & Gemini." https://shakalakaa.my/tools/ai-api-cost-calculator. Updated 2026-10-15. Licensed under CC BY 4.0.