AI AI Prices
AI API Cost Calculator Estimate monthly AI API costs Cheaper alternatives

AI API Cost Calculator

Estimate monthly AI API costs from provider, model, requests per day, average input tokens, average output tokens, cached ratio, and days per month. Start from the defaults to see daily, monthly, and yearly cost immediately.

View model list Need side-by-side comparison?
Estimate monthly AI API costs Default workload is already calculated Adjust provider, model, requests per day, average tokens, cached ratio, or days per month to replace the example with your own workload.
Use compare when The workload is fixed but the winning model is still unknown Compare is better for ranking several candidate models once your calculator inputs are stable.

If you searched for an AI API cost calculator, start here: the calculator turns token prices into business spend before showing price-table details.

Open compare with this scenario Review methodology

Scenario templates

Start from a realistic workload instead of a blank token form. Templates only preload values, so you can still edit everything after loading.

Clear template
Template

Article generation

Load template

Long-form writing with moderately sized prompts and substantial output.

4k input · 2.5k output · 1.5k requests/day · 10% cache Batch off
Template

Customer support

Load template

Short multi-turn support replies with high daily traffic and reusable system prompts.

1.8k input · 600 output · 120k requests/day · 35% cache Batch off
Template

RAG Q&A

Load template

Retrieval-heavy prompts where context dominates cost and cache can matter.

6k input · 800 output · 30k requests/day · 45% cache Batch off
Template

Code completion

Load template

Interactive coding assistance with medium prompts and short outputs.

2.5k input · 350 output · 50k requests/day · 20% cache Batch off
Template

Batch summarization

Load template

Offline bulk summarization where batch mode and cache reuse are both realistic.

12k input · 1.6k output · 8k requests/day · 50% cache · batch on Batch on

Estimate monthly AI API costs

Provider and model

3 selected. Leave this blank and the calculator falls back to three live-snapshot defaults.

Display currency

Keep workload inputs in focus with common switches first. Open the full currency search only when you need a less common display.

USD Default USD display
Common currencies
Search more currencies 165 supported currencies

Live FX rates come from Frankfurter. If conversion is unavailable, the calculator falls back to the model source currency and marks that row.

Quick token estimate

Optional helper for rough sizing before you set request token numbers.

Reset

Current workload summary

Scenario template Custom workload Directly edited inputs with no template preset.
Selected models 3
Display currency USD
Monthly request volume 300,000
Request shape 2,000 in · 1,000 out
Cache and batch 20% cache · Batch off
Budget Not set
FX mode Source currency

Default estimate result

Current winner: OpenAI · GPT-5 Nano

Daily Cost $4.82 10,000 requests per day
Monthly Cost $144.60 30 days per month
Yearly Cost $1735.20 12 months at the same workload assumptions

Cost Breakdown

Input Cost $0.0001 Uncached input share per request after 20% cached ratio
Output Cost $0.0004 1,000 average output tokens
Cached Input Cost $0.0000 120,000,000 cached input tokens per month
Cached Input Savings $5.40 Monthly reduction from cached input pricing when the provider lists it

GPT-5 Nano is the lowest monthly-cost option at USD 144.6, saving USD 116.4 per month (44.6%) versus GPT-4o Mini.

Cheaper Alternatives

Cheaper reason Same workload, lower monthly estimate Compared across 3 selected models, sorted by monthly cost low to high.

Request details

Show request details 
{
  "modelCodes": ["gpt-4o-mini", "gpt-4o", "gpt-5-nano"],
  "inputTokens": 2000,
  "outputTokens": 1000,
  "dailyRequests": 10000,
  "activeDays": 30,
  "cacheHitRatio": 0.20,
  "useBatch": false,
  "monthlyBudget": null,
  "displayCurrencyCode": "USD"
}

Token estimate helper

Useful when you know document size before you know request tokens.

From English words 0
From Chinese chars 0
From pages 0
Estimated total tokens 0

Estimate only. Default assumptions: 1 token ~= 0.75 English words, 1 token ~= 1.5 Chinese characters, 1 page ~= 500 English words.

How the calculator interprets inputs

  • `Cached ratio` discounts only the cached share of input tokens.
  • `Batch discount` applies only when the stored snapshot lists a batch ratio for that model.
  • `Budget fit` means the maximum monthly requests you can afford with the exact request shape above.
Open glossary Methodology

Share and export

Scenario, workload inputs, selected models, display currency, and budget stay in the current query string. You can share this exact AI API Cost Calculator state by URL or export the current estimate as CSV.

Export CSV

Estimated results

Monthly cost stays sorted low to high. Daily, monthly, yearly, and breakdown fields answer the workload cost before price-table details.

Model Daily cost Monthly cost Yearly cost Delta vs winner Cost breakdown 1k requests Blend / 1M Savings / month Budget status Budget fit
GPT-5 Nano
gpt-5-nano
OpenAI
Updated Jun 1, 22:05
Fallback source
PricePerToken OpenAI
Currency USD
Fallback snapshot from PricePerToken because OpenAI official pricing pages currently return an anti-bot challenge to server-side crawlers. Source updated at 2026-06-01T08:31:54.240890Z.
Cache saves $5.40 No batch savings
$4.82 $144.60 $1735.20
Current winner
Lowest monthly cost in this scenario
 Input $0.0001
Output $0.0004
Cached input $0.0000
 $0.482 $0.138 $5.40
cache + batch combined
Set budget or use a comparable display currency
Set budget to see fit
GPT-4o Mini
gpt-4o-mini
OpenAI
Updated Jun 1, 22:05
Fallback source
PricePerToken OpenAI
Currency USD
Fallback snapshot from PricePerToken because OpenAI official pricing pages currently return an anti-bot challenge to server-side crawlers. Source updated at 2026-06-01T08:31:54.241240Z.
Cache saves $9.00 No batch savings
$8.70 $261.00 $3132.00
$116.40
more per month than the winner
 Input $0.0002
Output $0.0006
Cached input $0.0000
 $0.870 $0.263 $9.00
cache + batch combined
Set budget or use a comparable display currency
Set budget to see fit
GPT-4o
gpt-4o
OpenAI
Updated Jun 1, 22:05
Fallback source
PricePerToken OpenAI
Currency USD
Fallback snapshot from PricePerToken because OpenAI official pricing pages currently return an anti-bot challenge to server-side crawlers. Source updated at 2026-06-01T08:31:54.241391Z.
Cache saves $150.00 No batch savings
$145.00 $4350.00 $52200.00
$4205.40
more per month than the winner
 Input $0.0040
Output $0.010
Cached input $0.0005
 $14.50 $4.38 $150.00
cache + batch combined
Set budget or use a comparable display currency
Set budget to see fit

Estimate only. Actual billing may differ by tokenizer behavior, cache hit rate, and provider rules.