Guides
Cost Tracking
The estimateCost utility maps token usage to estimated USD costs using a built-in price table for ~25 popular models.
Basic Usage
Code
import { estimateCost } from 'tekimax-ts'
const result = await client.text.generate({
model: 'gpt-4o',
messages: [{ role: 'user', content: 'Hello!' }]
})
const cost = estimateCost(result.usage, 'gpt-4o')
if (cost) {
console.log(`Input: $${cost.inputCost.toFixed(6)}`)
console.log(`Output: $${cost.outputCost.toFixed(6)}`)
console.log(`Total: $${cost.totalCost.toFixed(6)}`)
}Supported Models
Pricing is included for OpenAI (gpt-4o, gpt-4o-mini, gpt-4-turbo, o1, o3-mini), Anthropic (claude-3-5-sonnet, claude-3-opus, claude-3-haiku), Gemini (gemini-1.5-pro, gemini-1.5-flash, gemini-2.0-flash), Grok (grok-2), and popular OpenRouter models (deepseek-r1, llama-3-70b).
Use getSupportedCostModels() to list all models in the price table:
Code
import { getSupportedCostModels } from 'tekimax-ts'
console.log(getSupportedCostModels())
// ['gpt-4o', 'gpt-4o-mini', 'claude-3-5-sonnet-latest', ...]Custom Models
Register your own models or override existing prices:
Code
import { registerModelPrice } from 'tekimax-ts'
registerModelPrice('my-fine-tuned-model', {
inputPerMillion: 3.00,
outputPerMillion: 6.00,
})
// Now estimateCost works with your model
const cost = estimateCost(result.usage, 'my-fine-tuned-model')Cumulative Tracking
Track spending across multiple calls:
Code
let totalSpend = 0
for (const prompt of prompts) {
const result = await client.text.generate({ model: 'gpt-4o', messages: [{ role: 'user', content: prompt }] })
const cost = estimateCost(result.usage, 'gpt-4o')
if (cost) totalSpend += cost.totalCost
}
console.log(`Session total: $${totalSpend.toFixed(4)}`)Notes
- Returns
nullfor unknown models — this is intentional. It avoids throwing when you use a model that isn't in the price table. Check the return value. - Prices are estimates — they reflect published list prices and may not account for volume discounts, cached tokens, or provider-specific billing variations.
- Prices are per 1M tokens — the function does the math internally using
usage.promptTokensandusage.completionTokens.