Pricing Tables

NeuraMeter includes built-in pricing data for popular LLM models. Prices are used to calculate costMicrodollars automatically.

How It Works


import { getModelPricing, calculateCostMicrodollars } from '@neurameter/core';
 
const pricing = getModelPricing('openai', 'gpt-4o');
// { inputPricePerMToken: 2_500_000, outputPricePerMToken: 10_000_000, ... }
 
const cost = calculateCostMicrodollars(
  { inputTokens: 1000, outputTokens: 500 },
  pricing
);
// 7500 microdollars = $0.0075

Microdollar Arithmetic

All costs are stored as microdollars (integer arithmetic):


$1.00 = 1,000,000 microdollars
$0.01 = 10,000 microdollars

This avoids floating-point rounding errors common with dollar amounts.

Model Matching

getModelPricing(provider, model) supports:

Exact match — 'gpt-4o' matches 'gpt-4o'
Prefix match — 'gpt-4o-mini-2024-07-18' matches 'gpt-4o-mini'

OpenAI

Model	Input ($/1M)	Output ($/1M)	Cached Input ($/1M)	Reasoning ($/1M)
`gpt-4o`	$2.50	$10.00	$1.25	—
`gpt-4o-mini`	$0.15	$0.60	$0.075	—
`gpt-4.1`	$2.00	$8.00	$0.50	—
`gpt-4.1-mini`	$0.40	$1.60	$0.10	—
`o1`	$15.00	$60.00	$7.50	$60.00
`o3-mini`	$1.10	$4.40	$0.55	$4.40

Anthropic

Model	Input ($/1M)	Output ($/1M)	Cached Input ($/1M)
`claude-sonnet-4`	$3.00	$15.00	$0.30
`claude-haiku-4.5`	$0.80	$4.00	$0.08
`claude-opus-4`	$15.00	$75.00	$1.50

Google

Model	Input ($/1M)	Output ($/1M)	Cached Input ($/1M)
`gemini-2.5-pro`	$1.25	$10.00	$0.315
`gemini-2.5-flash`	$0.15	$0.60	$0.0375
`gemini-2.0-flash`	$0.10	$0.40	$0.025
`gemini-1.5-pro`	$1.25	$5.00	$0.315
`gemini-1.5-flash`	$0.075	$0.30	$0.01875

Groq

Model	Input ($/1M)	Output ($/1M)
`llama-3.3-70b-versatile`	$0.59	$0.79
`llama-3.1-8b-instant`	$0.05	$0.08
`llama-3-70b`	$0.59	$0.79
`mixtral-8x7b`	$0.24	$0.24
`gemma2-9b-it`	$0.20	$0.20

Mistral

Model	Input ($/1M)	Output ($/1M)
`mistral-large`	$2.00	$6.00
`mistral-small`	$0.20	$0.60
`codestral`	$0.30	$0.90
`mistral-embed`	$0.10	—
`open-mistral-nemo`	$0.15	$0.15

Custom Pricing

If a model is not in the built-in tables, getModelPricing() returns undefined and costMicrodollars defaults to 0. You can calculate cost manually:


const pricing = getModelPricing('openai', 'my-custom-model');
 
if (!pricing) {
  // Calculate manually
  const customPricing = {
    inputPricePerMToken: 1_000_000,    // $1.00 per 1M tokens
    outputPricePerMToken: 3_000_000,   // $3.00 per 1M tokens
  };
  const cost = calculateCostMicrodollars(usage, customPricing);
}

Updating Prices

Pricing data is bundled in the @neurameter/core package. To get updated prices, update the package:


npm update @neurameter/core