How does NumStack calculate AI token costs?

NumStack uses real published pricing from OpenAI, Anthropic, Google, and other AI providers. Enter your typical usage (input tokens, output tokens, requests per day) to see projected monthly costs across all models.

Why do different AI models have different token prices?

Different models have different computational costs. Smaller models like GPT-4o mini or Haiku are cheaper but less capable. Larger models like GPT-4o or Claude Sonnet are more powerful but cost more per token.

How can I reduce my AI API costs?

Use smaller models for simple tasks, cache repeated prompts, trim output length (output tokens cost 3-5x more than input), and batch requests instead of real-time when latency allows.

← Back to NumStack

API Documentation

Calculate AI token costs programmatically. Free tier needs no API key. Upgrade to Pro for 1,000 calls/day.

Base URL

url

https://numstack.net

Authentication

Free tier requires no authentication. Pro/Enterprise: pass your API key via header.

http

X-API-Key: ns_pro_YOUR_KEY_HERE
# or
Authorization: Bearer ns_pro_YOUR_KEY_HERE

🔑 Get a Pro API key at /pricing

Rate Limits

Tier	Daily Limit	Reset	Price
Free	10 calls	Midnight UTC	Free
Pro	1,000 calls	Midnight UTC	$4/month
Enterprise	Unlimited	—	Custom

Rate limit headers returned on every response: X-RateLimit-Limit, X-RateLimit-Remaining, X-RateLimit-Reset

Endpoints

POST/api/calculators/token-cost

Calculate token costs for a single model or compare all models.

Request Body

json

{
  "model": "gpt-4o",          // Model ID — required unless compareAll: true
  "inputTokens": 1000,        // Number of input tokens per request — required
  "outputTokens": 500,        // Number of output tokens per request — required
  "requestsPerDay": 100,      // Requests per day for projections — default: 1
  "compareAll": false         // Set true to compare all models — ignores "model"
}

Single Model Example

bash

curl -X POST https://numstack.net/api/calculators/token-cost \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "inputTokens": 1000,
    "outputTokens": 500,
    "requestsPerDay": 100
  }'

Response

json

{
  "model": "gpt-4o",
  "name": "GPT-4o",
  "provider": "OpenAI",
  "query": { "inputTokens": 1000, "outputTokens": 500, "requestsPerDay": 100 },
  "costs": {
    "perRequest": 0.0075,
    "perDay": 0.75,
    "perWeek": 5.25,
    "perMonth": 22.5,
    "perYear": 273.75
  },
  "pricing": {
    "input_per_1m_tokens": 2.5,
    "output_per_1m_tokens": 10
  },
  "currency": "USD",
  "rateLimit": { "remaining": 9, "limit": 10, "resetAt": "2026-05-05T00:00:00.000Z" },
  "meta": {
    "calculator": "token-cost",
    "version": "1.0",
    "tier": "free",
    "pricingUpdated": "April 23, 2026",
    "source": "numstack.net"
  }
}

Compare All Models

bash

curl -X POST https://numstack.net/api/calculators/token-cost \
  -H "Content-Type: application/json" \
  -d '{
    "compareAll": true,
    "inputTokens": 1000,
    "outputTokens": 500,
    "requestsPerDay": 100
  }'

With Pro API Key

bash

curl -X POST https://numstack.net/api/calculators/token-cost \
  -H "Content-Type: application/json" \
  -H "X-API-Key: ns_pro_YOUR_KEY_HERE" \
  -d '{ "model": "claude-sonnet-4", "inputTokens": 2000, "outputTokens": 1000 }'

GET/api/v1/models

List all supported models and their pricing.

bash

curl https://numstack.net/api/v1/models

JavaScript / Node.js

javascript

const res = await fetch('https://numstack.net/api/calculators/token-cost', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'X-API-Key': process.env.NUMSTACK_API_KEY, // optional, for Pro tier
  },
  body: JSON.stringify({
    model: 'gpt-4o',
    inputTokens: 1000,
    outputTokens: 500,
    requestsPerDay: 100,
  }),
});
const data = await res.json();
console.log(`Monthly cost: $${data.costs.perMonth}`);

Error Codes

Status	Meaning
400	Bad Request — missing or invalid parameters
403	Forbidden — calculator not available on your plan
404	Model Not Found — check /api/v1/models for valid IDs
429	Rate Limited — daily limit exceeded, see X-RateLimit-Reset
500	Server Error — please report to [email protected]

Need more than 10 calls/day?

Pro gives you 1,000 calls/day, all calculators, no ads, and export. $4/month.

See Pricing →