pricetoken

Real-time LLM pricing data. Typed client, offline cost calculator, and static pricing for every major provider.

Install

npm install pricetoken

Quick start

import { PriceTokenClient } from 'pricetoken';

const pt = new PriceTokenClient();

// Get all model pricing
const models = await pt.getPricing();

// Filter by provider
const anthropic = await pt.getPricing({ provider: 'anthropic' });

// Get a single model
const opus = await pt.getModel('claude-opus-4-6');

// Find the cheapest model
const cheapest = await pt.getCheapest();

// Compare models side by side
const compared = await pt.compare(['gpt-4o', 'claude-sonnet-4-6']);

// Price history (last 30 days)
const history = await pt.getHistory({ days: 30 });

// List providers
const providers = await pt.getProviders();

Authentication

Free tier: 30 requests/hour (no key required).

For higher limits, pass an API key:

const pt = new PriceTokenClient({ apiKey: 'pt_...' });

Offline cost calculation

Calculate costs without any API calls using bundled static pricing:

import { calculateModelCost } from 'pricetoken';

const cost = calculateModelCost('gpt-4o', 1000, 500);
// { modelId: 'gpt-4o', inputTokens: 1000, outputTokens: 500,
//   inputCost: 0.0025, outputCost: 0.0075, totalCost: 0.01 }

Or use your own pricing data:

import { calculateCost } from 'pricetoken';

const cost = calculateCost('my-model', 2.50, 10.00, 1000, 500);

Static pricing data

Access bundled pricing without network calls:

import { STATIC_PRICING } from 'pricetoken';

console.log(STATIC_PRICING.length); // 21 models

Note: Static data reflects prices at the time the package was published and does not auto-update. For the latest prices, use PriceTokenClient to fetch live data from the API.

API

PriceTokenClient

| Method | Returns | Description | |--------|---------|-------------| | getPricing(opts?) | ModelPricing[] | All models. Filter by provider, currency. | | getModel(modelId, opts?) | ModelPricing | Single model by ID. | | getCheapest(opts?) | ModelPricing | Cheapest model. Filter by provider. | | compare(modelIds, opts?) | ModelPricing[] | Compare specific models. | | getHistory(opts?) | ModelHistory[] | Price history. Filter by days, modelId, provider. | | getProviders() | ProviderSummary[] | All providers with model counts. |

Options

new PriceTokenClient({
  baseUrl: 'https://pricetoken.ai', // default
  apiKey: 'pt_...',                  // optional
});

Types

All types are exported:

import type {
  ModelPricing,
  PriceHistoryPoint,
  ModelHistory,
  ProviderSummary,
  CostEstimate,
  PriceTokenResponse,
  PriceTokenError,
} from 'pricetoken';

Disclaimer

Pricing data is provided on a best-effort basis and may be inaccurate, incomplete, or outdated. LLM providers change prices without notice, and our scraping pipeline may not capture every change immediately.

This data is for informational purposes only. Do not use it as the sole basis for financial decisions. Always verify pricing directly with the provider before committing to spend.

If you get a bill you weren't expecting, that's between you and your provider — not us. See the MIT License under which this project is distributed (specifically the "AS IS" and "NO WARRANTY" clauses).

Found incorrect pricing? Open an issue.

License

MIT